:00
=co
= CO
•CM
loo
CO
WIDSITH
CAMBRIDGE UNIVERSITY PRESS
Eonion: FETTER LANE, E.G.
C. F. CLAY, MANAGER
Ototnburgfj: too, PRINCES STREET
Berlin: A, ASHER AND CO.
Irfpjtfl: F. A. BROCKHAUS
$eto Hork: G. P. PUTNAM'S SONS
an» (ZTatcutta: MACMILLAN AND CO., LTD.
All rights reserved
WIDSITH
A STUDY IN OLD ENGLISH
HEROIC LEGEND
BY
trt^ -
R. W. CHAMBERS, M.A.
FELLOW AND LIBRARIAN OF UNIVERSITY COLLEGE, LONDON
Cambridge :
at the University Press
1912
1792
Cs
Cambrtfige:
PRINTED BY JOHN CLAY, M.A.
AT THE UNIVERSITY PRESS
TO
W. P. KER
PREFACE
increasing interest shown during the last thirty years
J- in Beowulf has, in England at least, not been extended
to the other monuments of Old English heroic poetry : and that
poetry has often, in consequence, been somewhat misjudged.
The chief merit of Widsith is that it enables us to make a more
correct estimate.
If Widsith had been, as some have thought it, an authentic
record of a visit to the court of Ermanaric, it would have been
valuable : though hardly so valuable as the account left by
Priscus of his visit to the court of Attila, an account which
nevertheless few trouble to read. But Widsith is more to us
than any record of fourth or fifth century travel could have
been : it is a record of lost heroic song. As one of its earliest
editors said, with sound sense, if in questionable Latin, Discimus
ex eo, quot carmina temporum injurid nos amisimus.
Another advantage to be drawn from a study of Widsith is
that here we find the older scholars at their best. It is remark
able, considering the means at his disposal, how good are the
comments even of the despised Conybeare, and still more is
this true of those of Kemble, Leo, Lappenberg, Ettmiiller, and
in more recent days, of Moller, Ten Brink, and Miillenhoff.
I hope that I have shown, both in dealing with the Harlung
story and the Offa story, that my reverence for Miillenhoff is
not a superstitious one. But it is time to protest against the
undue depreciation of that great scholar, which has of late been
prevalent, particularly in England and America. This deprecia
tion is often not according to knowledge : indeed, we find later
viii Preface
students, even the most scholarly, either ignorant of facts which
were perfectly familiar to Kemble and Miillenhoff, or else hailing
such facts as new discoveries.
Among the more recent students, I have to acknowledge
indebtedness most frequently to Binz, Brandl, Bremer, Chad-
wick, Heusler, Holthausen, Heinzel, Jiriczek, Lawrence, Much,
Olrik, Panzer, Symons. I have tried to express these and other
obligations in the due place. Where there was general agree
ment on any point, I have been able to pass over it the more
rapidly : where I had to differ from an opinion which has heavy
critical authority behind it, I have felt bound to give my reasons
at the greatest length space would permit. I regret that this
tends to concentrate attention unduly upon points of difference.
The more the " literature of the subject " is studied, the less
room does there seem to be for new views ; over and over again
I have discovered that some observation which I was hoping to
find new had been anticipated — by Brandl, Panzer, Olrik,
Heinzel, or Rajna. But instances probably remain in which
a view which has already been expressed by someone else is
put forward as my own. I cannot hope to have run to earth
every discussion of every hero or tribe recorded in Widsith.
And indeed, much has been published, especially during the
past fifteen years, and in journals of high repute, which it is
hardly necessary to put on record. I have therefore reserved
the right of omitting references to comments which did not
appear to lead us further : other omissions are no doubt the
result of oversight. For the same reason, in the section on
grammatical forms I have neglected spellings like Moidum
for Medum (cf. Oiddi : Eddi) or Amothingum (th for ]?) which,
though interesting, could hardly be made the basis or support
of any argument : and, in dealing with the metrical effect of the
syncope of u, I have neglected lines like On }>am sieoc hund wees,
where the loss of an earlier u might easily have been concealed
by a slight rearrangement of the words.
Finally I have the pleasure of acknowledging the help in
reading proofs very kindly given by Miss M. Eyre, Miss E. V.
Preface ix
Hitchcock, Mr L. A. Magnus, and, I should add, by the
Cambridge Press reader. To Mr J. H. G. Grattan I am in
debted for many corrections and suggestions, and to Professor
Robert Priebsch for a constant advice which has been every
where of service, but more particularly in the sections dealing
with Ermanaric, Hama, and Heoden ; to Professor W. P. Ker
my debt is even heavier, and more difficult to define ; it will be
found, like the blow which Offa gave his adversary, non unam
partem sed totam transegisse compagem.
R. W. CHAMBERS.
31 December 1911.
CONTENTS
CHAP. PAGE
I. WlDBITH AND THE GERMAN HEROIC AGE .... 1
II. THE STORIES KNOWN TO WwsiTB: GOTHIC AND BDRGUNDIAN
HEROES 12
III. THE STORIES KNOWN TO WlDSITH: TALES OF THE SEA-FOLK,
OF THE FRANKS AND OF THE LOMBARDS ... 66
IV. WlDSITH AND THE CRITICS 127
V. THE GEOGRAPHY OF WJDSITH 153
VI. THE LANGUAGE AND METRE OF WIDSITH . . . .166
VII. SUMMARY AND CONCLUSION 177
TEXT OF WIDSITB, WITH NOTES 187
APPENDIX :
(A) Bibliography . .225
(B) " Maurungani" and the " Geographer of Ravenna " 235
(C) Eastgota 236
(D) The Jutes 237
(E) The original homes of the Angli and Varini
(Engle and Wsernas) 241
(F) Iste 248
(G) Idumingas . . 250
(H) The term Hrcedas applied to the Goths . . 252
(I) Ermanaric as the foe of the Huns . . . 253
(K) The last English allusion to Wudga and Hama 254
(L) The Geography of the Orosius compared with that
of Widsith 254
(M) Schiitte's Law of initial and terminal stress . 255
(N) The beag given by Eormanric .... 256
MAPS AND INDEX . ... . 258
CORRECTIONS AND ADDITIONAL NOTES
p. 7. For a parallel to the ' Biblical interpolation ' in Widsith see Matthew
Arnold ' On the study of Celtic Literature,' p. 69 (1867 edit.).
p. 25. For ' common grandfather Hrethel ' read ' kinsman Hrethel. '
p. 172. The assertion that 1. 45 a, if we substitute for Hrof&gar the earlier
Hroftgaru, can still be paralleled in Beowulf, needs qualification. For in half-
lines like sellice sadracan, fyrdsearu fuslicu, the syllable lie should probably be
read short (cf. Sievers, P.B.B. x, 504; xxix, 568). Yet, for the other reasons
given on p. 172, 1 do not think that Hroftgaru is impossible : and even if it were,
there are three reasons, any one of which would prevent our basing upon this
line an argument for dating the bulk of Widsith later than 700 A.D. For, firstly,
11. 45 etc. are presumably a subsequent addition to the Catalogue of Kings (see
pp. 135, 147) : secondly, a simple rearrangement of the words would make the
line run satisfactorily : and thirdly, after a long syllable, bearing the subsidiary
accent, u was lost earlier than after a long syllable bearing the main accent, and
was certainly already lost before 700.
CHAPTER I.
WIDSITH AND THE GERMAN HEROIC AGE.
The Heroic Poetry of the Germans.
OVER many great races an enthusiastic movement seems at
a certain period to sweep, carrying them during a few years to
success, alike in arms and song, till the- stream sinks back into
its old channel, and the nation continues a career, honourable,
it may be, but wanting in the peculiar ardour of its great age.
Such a spirit as came upon the Athens of Pericles, or the
England of Elizabeth, seems to have animated the widely
scattered Germanic tribes which, in the fifth and sixth centuries,
plundered and drank and sang amid the ruins of imperial
Rome.
To the cultured Roman provincial, trying to lead an elegant
and lettered existence amid reminiscences of the great ages of
Classical history, the drinking seemed immoderate, and the song
dissonant. One such has told us how his soul was vexed at the
barbarous songs of his long-haired Burgundian neighbours, how
he had to suppress his disgust, and praise these German lays,
though with a wry face. How gladly now would we give all
his verses for ten lines of the songs in which these " long-haired,
seven foot high, onion-eating barbarians " celebrated, it may be,
the open-handed ness of Gibica, or perhaps told how, in that
last terrible battle, their fathers had fallen fighting around
o. 1
2 Widsith
Gundahari1. But the Roman, far from caring to record these
things, was anxious to shut his ears to them. The orthodox
bishop of the fifth century, when compelled to listen to the
songs of German tribesmen, seems to agree with the apostate
emperor of the fourth, who thought these lays like the croaking
of harsh-voiced birds2. So it has come to pass that, whilst
the contemporary Latin verse has been only too abundantly
preserved, the songs chanted at the feasts of Gothic and
Burgundian, Frankish and Lombard, Danish and Anglian
chiefs, in the fifth and sixth centuries, have been lost. The
loss is among the most lamentable in the history of poetry.
Yet a few records have come down to us, for, as during
centuries these songs passed, in the mouths of minstrels, from
one branch to another of the Teutonic race8, they often came to
be written down. In later days Charles the Great caused the
Frankish versions to be collected4. Upon the English poems
Alfred the Great was educated5 : upon them in turn he
educated his children6; and he was wont to recommend to
others the learning of " Saxon " lays by heart7. In this case,
too, our authority makes mention of written books. It is
1 Quid me, etsi valeam, parare carmen
Fescenninicolffi jubes Diones
inter crinigeras situm catervas
et Germanica verba sustinentem,
laudantem tetrico subinde vultu
quod Burgundio cantat esculentus,
infundens acido comam butyro?
vis dicam tibi quid poema frangat?
ex hoc barbaricis abacta plectris
spernit senipedem stilum Thalia
ex quo septipedes videt patronos, &c.
Sidonius Apollinaris, Car. xn, ed. Luetjohann,
p. 231 (in M.G.H.).
2 £0ea(r<i/J.T)v roi KO.I TOVS virtp rbv 'Prjvov Papfidpovs dypia fi^Xrj \££et. ireiroiyfdva,
irapair\7]<Ttq. rols Kptajfj-ois TWV rpa\v ftoibvrwv 6pvL6iav frSovTas ical eti<f>paivofi,frovs 4irl
rois fj^\effiv. (MiffOTTuyuv : Julian, ed. Hertlein, 1875-6, p. 434.)
3 There is some evidence that not merely traditions, but complete lays
passed from one Germanic nation to another. (Cf. Heusler, L.u.E. 4 ; Symons
in Z.f.d.Ph. xxxvm, 164.)
* Einhard, Vita Carol, c. 29. In Pertz (fol.) SS. n, 458.
5 Asser, ed Stevenson, 1904, § 22 Sed Saxonica poemata die noctuque solers
auditor, relatu aliorum saepissime audiens, docibilis memoriterretinebat...23 Cum
ergo quodam die mater sua...quendam Saxonicum poematicae artis librum...
ostenderet...
6 § 75 Eadwerd et j33lfthryth...et psalmos et Saxonicos libros et maxime
Saxonica carmina studiose didicere.
7 § 76 Interea tamen rex...et Saxonicos libros recitare, et maxime carmina
Saxonica memoriter discere aliis imperare , , .non desinebat.
Widsith and the German heroic age 3
likely, however, that, more often than not, these Germanic lays
were not committed to writing at all. Of those that were,
England has preserved a few, fragmentary but most precious,
and continental Germany one.
Yet we can form but a poor idea of this old literature from
the surviving fragments of the Songs of Waldere or Hildebrand,
still less from Beoiuulf, the one poem which happens to have
come down to us complete, but of which the plot is singularly
poor. A better estimate of the value of these vanished songs
can be made, partly from certain passages in the Latin
chroniclers which are obviously founded upon lost lays, partly
from the magnificent afterglow of the Old Norse heroic verse
and prose saga; and not least from the glamour which an
English poet, using the old motives and the old common
places, has been able to throw over the historic facts of the
death, in battle against the heathen, of an Essex nobleman in
the days of Ethelred the Unready.
So this world of high-spirited, chivalrous song has passed
away, leaving fragments so scanty that it requires some sym
pathetic allowance before we can realise how great the loss has
been :
hu seo >rag gewat
genap under nihthelm, swa heo no waere !
We cannot accept as compensation the remodelled stories in
the thirteenth century High German garb in which they have
come down to us. Sifrit, the horny-skinned wife-beater, Gunther
hanging from a peg in his bedchamber1, are but poor substitutes
for the lost figures of the old Teutonic hero-song, Gundahari
the Burgundian, Offa the Angle, Ermanaric the tyrant or the
noble Theodoric. The Nibelungen Lied is beyond dispute one
of the world's greatest things. Inferior as it is in detail to the
Greek epic, yet, when all is drawing to an end, and the terrible
Hagene has been overcome by the perfect knight, Dietrich von
Bern, we are left with a deep awe akin to what we feel when
we see Priam at the knees of Achilles. The Nibelungen Lied
is glorious, but its glory, like its metre, is not that of the
ancient lays.
1 ,Der Nibelunge Not (Lachmann, 590*, 837* ; Bartsph, 640, 894).
1—2
4 Widsith
It is our duty then to gather up reverently such fragments
of the old Teutonic epic as fortune has preserved in our English
tongue, and to learn from them all we can of that collection
of stories of which these fragments are the earliest vernacular
record
Widsith.
The weightiest vernacular document bearing upon the old
German hero-cycles is a poem found in the Exeter Book, that
collection of English religious verse, transcribed in the late
tenth or early eleventh century, which was given to his cathedral
church by Leofric, first Bishop of Exeter (died 1072).
Widsith and Deor, which stand respectively eleventh and
twentieth in the collection, differ remarkably from the other
poems in the volume. They are relics of heathendom embedded
in a Christian anthology. Both are lyrics founded upon epic
tradition : in each a glee man tells a tale of himself, adorning
his song by copious allusions to the stories he loves.
Widsith — "Farway" — the ideal wandering minstrel, tells of
all the tribes among which he has sojourned, all the chieftains
he has known. The first English students1 of the poem re
garded it as autobiographical, as the actual record of his
wanderings written by a real scdp', and were inclined to
dismiss as interpolations passages mentioning princes whom
it was chronologically impossible for a man who had met
Ermanaric to have known. This view was reduced to an
absurdity by Haigh, who pointed out that a certain hero,
Hama, is mentioned last. But what reason can have led to
his being mentioned last ? Clearly only one ; the Traveller
1 For example : Conybeare, Illustrations, 1826, p. 28 : Kemble, Preface to
Beowulf, 1833, p. xvii : Guest, Rhythms, 1838, vol. n, pp. 76, etc.
Stopford Brooke (English Literature from the beginning to the Norman
Conquest, 1898, pp. 46-8) and Garnett (English Literature, i, pp. 7-8) have,
more recently, been inclined to regard the poem in this light, though with
considerable hesitation; and so, with perhaps less hesitation, have Anderson
(pp. 5, 23) and Chadwick (Camb. Hist, i, 36).
Brandl, too, speaks of Widsith as autobiographical (Berl. Sitzungsberichte,
xxxv, 1908); a view which is disputed by Schiicking (Engl. Stud. XLII, 111).
But I take Brandl to mean merely that Widsith is cast in an autobiographical
form, not that the travels narrated are historic; this last view would not be
consistent with what Brandl has said elsewhere about Widsith (Pauls Grdr. M
n, 1, 968).
Widsith and the German heroic age 5
himself, Haigh urged, was Hama, and was moved by motives of
modesty \
The temptation to attribute historic value to poetry in
which the names of historic chiefs often meet us is, of course,
strong; and, giving way to it, the early chroniclers of many
nations have incorporated heroic tradition into their histories.
But it is an essential characteristic of heroic poetry that, whilst
it preserves many historic names, it gives the story modified
almost past recognition by generations of poetic tradition.
Accurate chronology too is, in the absence of written records, ;
impossible : all the great historic chieftains become con- ;
temporaries : their deeds are confused : only their names, and
sometimes their characters, remain.
The more we study the growth of German heroic tradition,
the more clear does it become that Widsith and Deor reflect
that tradition. They are not the outpourings of actual poets
at the court of Ermanaric or the Heodenings. What the poems
sung in the hall of Ermanaric were like we shall never know :
but we can safely say that they were unlike Widsith. The mere
excision of a name here and there will not put our poem back
to the date of Ermanaric, any more than the excision of a few
late features will make Mallory's Morte d'Arthur a correct
account of the court of a sixth century British king2.
The reasons which prevent us from regarding Widsith as the
itinerary of a real scop have been stated fully and convincingly
by a recent critic8. To suppose that the author, in his own
person, claims to have come into contact with Ermanaric is
indeed to misunderstand the spirit of the poem. The Traveller's
tale is a phantasy of some man, keenly interested in the old
1 Haigh, Sagas, p. 148.
2 I cannot agree with Mr Chadwick's attempt to draw historic and even
chronological data from Widsith. See Origin of the English Nation, 135. How
limited is the historic element in German heroic tradition has been well shown
by Heusler, Berl. Sitzungsberichte, xxxvn, 1909.
3 Lawrence in Mod. Philol. rv, 355, etc., 367, etc. The careful and measured
statement of Dr Lawrence appears to me to be conclusive. But the view that
the Traveller's wanderings are at least in part historic is still so prevalent
among English scholars that I have thought it better to summarize below the
considerations which, as it appears to me, make it impossible so to regard
them, from an investigation first of heroic legend (pp. 142-3) and then of
geography (pp. 163-5).
6 Widsith
stories, who depicts an ideal wandering singer, and makes him
move hither and thither among the tribes and the heroes whose
stories he loves. In the names of its chiefs, in the names of its
tribes, but above all in its spirit, Widsith reflects the heroic age
of the migrations, an age which had hardly begun in the days
/ of Ermanaric.
Catalogues of this kind are peculiarly liable to interpolation,
especially during a period of oral tradition. There is a natural
tendency for each gleeman through whose hands the poem
passes to corrupt the names of clans and heroes with which
he is not familiar, and to add the names of others which his
tribal prejudices lead him to think should not have been left
without honourable mention. Hence catalogues, from those in
the second book of the Iliad or the eleventh book of the
Odyssey downwards, are always to be suspected1.
But in Widsith we have a catalogue of some seventy tribes2 ;
and of sixty-nine heroes, many of whom can be proved to have
existed in the third, fourth and fifth centuries of our era, and
the latest of whom belong to the sixth century. Yet, although
every chief whom we can date lived prior, not only to the con
version of the English to Christianity, but even to the com
pletion of their settlement in Britain, this thoroughly heathen
poem has come down to us in a transcript which some English
monk made about the year 1000. Clearly it will require
patience to disentangle its history, and fix its approximate
date and place of origin.
1 A comparison of Widsith with the Greek heroic catalogues would be
helpful: it has been indicated, without details, by Meyer (Altgermanische Poesie,
1889, p. 4). We might also compare later mediaeval lists of romances such as
those given in Skelton's Philip Sparrow, those quoted in the appendix to Ker's
Epic and Romance, or those Italian catalogues printed by Pio Eajna (Zeitschrift
/. Romanifche Philologie, n, 1878, pp. 138 etc., 222 etc.). The likeness of these
latter to Widsith was noted by Eajna himself (Origini, p. 40). But the closest
parallel of all is found in the Icelandic name lists, as given in the heroic muster
rolls, in the Thulor, and in poems like Grimnismdl, Hyndloljfy, the opening
lines of the Angantyr lay (cf. C.P.B. i, 565), or the enumeration of dwarfs in
Volospd. For later Icelandic lists see Kolbing, Beitrdge zur vergleichenden
Geschichte der romantischen Poesie, 1876, pp. 154-5, and C.P.B. i, pp. xviii, cxi.
2 Including twenty-two referred to only in the doubtful passage.
Widsith and the German heroic age
General Impressions of the Poem. Obvious Interpolations.
Age of the subject-matter.
Whilst the general consensus of opinion would make our
poem one of the oldest, if not quite the oldest, in Germanic
literature, one critic of repute has, on the strength of certain
Biblical and Oriental names occurring in it, placed it as late as
the early ninth century1.
Such a view has, however, been justly condemned by
Miillenhoff as superficial2. It neglects the fact, so obvious
even to a casual observer, that these Biblical and Oriental
names are confined to one brief passage, and are alien to the
spirit of the poem as a whole. The author was clearly a man
steeped in all the lore of the Teutonic races : he makes his
Widsith visit Swedes and Burgundians, Angles and Danes,
Lombards and Goths : he is interested in the national heroes
of all these folk : he knows something too of the peoples across
the mark : Widsith has visited the Finns of the North and the
Wends of the East, has heard tell of the Caesar of the East
and has been in Italy. He is skilled in traditional and
genealogical lore, in those "old unhappy far off things and
battles long ago " which are so highly rated in the heroic age
of any people. But we may safely say of the poet that his
study was but little on the Bible. The student who reads
Widsith for the first time passes therefore, with a feeling of
surprise, from a list of tribes with whose stories any gleeman
of the heroic age would be familiar, to a series of supposititious
travels among Medes and Persians, Israelites and Hebrews.
Accordingly this passage was, at a very early stage in the
criticism of Widsith, condemned as spurious3; and for a critic,
1 K. Maurer in Zacher's Z.f.d.Ph. n (1870), 447.
2 D.A. v, 65-6.
3 By Kemble. " More than one circumstance, however, compels us to con
clude that later transcribers added to the work of Hermanaric's contemporary.
Not only is the insertion of several nations only known in Europe through the
Bible, and the distinction between Hebrews and Israelites suspicious, but the
mention of Theodoric the king of the Franks, if by this prince the son of
Chlodowic be meant, places the poem in which his name occurs later at
least than A.D. 480." Preface to Beowulf, 1833, p. xvii. Leo, Sprachproben, 82,
endorsed this view.
8 Wtdnth
in trying to fix the date of the poem, to overlook its doubtful
nature and to base his theories upon it, is, as Miillenhoff has
stated, uncritical.
It is true that the certainty with which a German scholar
will sometimes fix upon this or that passage as interpolated is
apt to arouse scepticism in English students1. But here the
evidence is peculiarly strong, and we are compelled to endorse
the judgment of Miillenhoff, that a critic will hardly recognise
any passage as interpolated, if he does not so recognise this2.
Exactly how long is the interpolated passage, it is more
difficult to say. The four lines of Oriental adventure (75, 82-4)
must certainly go :
Mid Sercingum ic waes and mid Seringum, ...
Mid Israhelum ic wees and mid Exsyringum,
Mid Ebreum and mid Indeum and mid Egyptum.
Mid Moid urn ic waes and mid Persum, and mid Myrgingum.
Miillenhoff is perhaps right in condemning the whole thirteen
lines, 75 — 87, and in including in the same condemnation lines
14 — 17 (Alexandreas) and lines 131 — 134. Our judgment must
depend upon our interpretation of certain puzzling individual
words. As an example of this, line 87 may be taken,
And mid Eolum and mid Istum and Idumingum.
What tribe can be signified under the name Eolum is unknown;
and if we are to attach any meaning to the word, we must alter
the text. Istum in all probability refers to the Esths, whatever
race may have been designated by that name, but our judgment
of the verse must depend upon the interpretation of Idumingum.
If these are the Idumaeans of the Bible, then the verse must
be attributed to our late monkish interpolator. But it is at
least possible that we have here a record of a Baltic tribe,
neighbours and near of kin to the Esths, the Ydumaei, whose
unruly doings are recorded by a Livonian chronicler of a later
period3. Tf this be so, we have in the juxtaposition of Esths
and Ydumaei a piece of geographical information of the kind
which the author or authors of the original poem would be
1 Cf. Jebb, Oedipus Coloneus, 1885, p. lii ; Morley's English Writers, u, 31.
2 Z.f.d.A. xi, 291.
3 See Appendix G, Idumingas.
Widsith and the German heroic age 9
likely to possess, but which would appear to be quite beyond
the reach of the tenth century English monk whom we suppose
to have interpolated lines 82 — 84; and we must in that case
pronounce the lines genuine.
If, however, we place on one side the four lines 72, 82 — 4
as interpolated, and the other lines rejected by Miillenhoff as
suspect, the poem leaves upon us a very definite impression.
It is a catalogue of the tribes and heroes of Germany, and
many of these heroes, though they may have been half legendary
already to the writer of the poem, are historic characters who
can be dated with accuracy. Widsith has been at the court of
the most powerful of all the Gothic kings, Eormanric, who
died A.D. 375 ; and of Guthhere (Gundahari) the Burgundian,
who ruled at Worms from 413 to 437. He knows of ^Etla
(Attila), who reigned from 433 to 453, and of the Frankish
Theodoric, who reigned from 511 to 534. He even mentions
being in Italy with ^Elfwine (Alboin) who invaded Italy in 568
and died in 572 or 573 \
On the other hand, scholars have, with few exceptions,
assumed that no mention is made of the Ostrogothic Theodoric,
the conqueror of Italy, who reigned from 474 to 526, and whose
fame was to eclipse that of all the chiefs just mentioned, till
Eormanric (Ermenrich), Guthhere (Gunther), and JEtla (Etzel)
came to be mere foils to the heroic figure of the mirror of
knighthood, Theodoric of Verona : Dietrich von Bern.
The assumption that Dietrich von Bern does not appear in
Widsith was made too hastily, as can be proved ; but certainly
the figure of the great Ostrogothic king had not yet assumed
its gigantic proportions when our poem was written. If
mentioned at all, Theodoric is thought of as a retainer of
Eormanric : later poets would have thought of Eormanric only
as the uncle and foe of Theodoric.
Our poem then reflects a definitely marked period : that of
those barbarian invasions of the Roman Empire, which began
when the Goths first swarmed across the frontier, and ended
with Alboin's falling upon an Italy worn by famine, pestilence
1 See Hodgkin, Italy, v, 168 (footnote).
10 Widsith
and the sword of Visigoth, Hun, Vandal and Ostrogoth.
Whether Widsith is to be regarded as the work of one man, or
as a cento of several heroic catalogues, need not for the moment
trouble us; for, though the various sections of the poem may
show minor idiosyncrasies, they all reflect the same heroic age.
Excluding the lines of Biblical lore, we are left with a poem
recording the tribes and chiefs of the German migrations, from
the middle of the third to the middle of the sixth century:
from Ostrogotha to Alboin.
Alboin gives us then a limit to the age of the poem in its
present form. It must be later than 568, not necessarily much
later, for we need not scrupulously allot a term of years to
allow of ^Elfwine receding into the legendary plane of the
earlier Eormanric and ^Etla: to the far-off dwellers on the
North Sea, both might perhaps soon seem equally distant.
How much later, we can only guess after a rather minute
examination of the stories of the chiefs and peoples recorded in
Widsith. If we find the poet constantly referring to a state
of things which had passed out of existence by the seventh
century, this, together with his silence on all matters later than
the year 568, will incline us to regard the matter of the poem,
approximately, as belonging to the last decades of the sixth
century or the first of the seventh.
In the sixth century each of the innumerable tribes into
which each of the five great branches of the German people
was subdivided was an independent nation, with its own laws,
traditions and policy ; and these nations were scattered over all
Europe and part of Northern Africa. No man could know
them all, and it is quite easy to see wherein the interests of
the poet of Widsith lay. That poem does not belie the English
dress in which it has come down to us. It is in the countries
bordering on the Baltic and the North Sea, be seem tweonum, that
we must locate the tales which interest him : the stories of
Sceaf, Hagena and Heoden, Finn Folcwalding, Offa the Angle,
Breca of the Brondings, Hrothgar and Ingeld.
But one branch of the German people had done and suffered
so much that their heroes became household names throughout
Widsith and the German heroic age 11
all Germania. The Gothic champions, Eastgota and Eormanric,
Wudga and Hama, and the Huns who, first as the enemies,
then as the allies, and again as the enemies of the Goths, are
inextricably entangled in their history and saga: these are
as familiar to the poet of Widsith as his own sea-heroes.
Widsith visits many people : but he starts from amid the
sea-folk, and his objective is the Gothic court. We shall find it
easy to get order out of the apparent chaos of his tradition and
ethnology, if we divide the stories he loves into (1) Gothic saga
and (2) the tales of the sea-folk.
At the outset, this is summed up in seven words when it is
said of the minstrel :
HreScyninges ham gesohte,
eastan of Ongle, Eormanrices.
NOTE. When a Germanic hero is spoken of without
reference to any particular version of his story, I give the
name in the current historical form ; but when referring
particularly to any version, in the form proper to that version.
Thus Ermanaric is the form generally used, Eormanric and
Jormunrekk being reserved respectively for the king of Old
English and of Scandinavian legend. Similarly Gundahari is
used for the historic king : O.E. Guthhere, O.N. Gunnar, M.H.G.
Gunther for the hero of particular traditions.
A difficulty arises in the case of heroes who have no known
historical prototype. Here the Old English name (e.g. Hama)
has to be used both generally, and also as the specifically
O.E. form ; a form like Heimir is left for the champion as
depicted in the Thidreks saga.
When Scandinavian names are transliterated S is repre
sented by th, not by d : v by v, not by w. But where a form
seems to be definitely established with d or w, it is retained :
Swanhild, Swein, Sigurd, Thidreks saga. The r of the nomina
tive is kept after vowels, but not after consonants : Heimir,
Jormunrekk, Th ith rek.
CHAPTEE II.
THE STORIES KNOWN TO WIDSITH: GOTHIC AND
BURGUNDIAN HEROES.
The beginnings of Heroic Song in Germany.
TACITUS1 mentions the ancient songs in which the Germans
celebrated Tuisco, the earth-sprung god, and his son Mannus,
the authors and founders of their race. Whether these poems
were narrative or choric we do not know. That not gods alone
but also historical and semi-historical heroes were com
memorated, is proved however by the further remark of
Tacitus concerning the old poems " quod unum apud illos
memoriae et annalium genus est." One other precious scrap
of information Tacitus has preserved. When making his last
mention of Arminius, and telling how Tiberius refused to
compass by poison the death of the enemy of the Roman
people, he sums up thus the further history of the liberator of
Germany: "dolo propinquorum cecidit . . . septem et triginta
annos vitas, duodecim potentia3 explevit, caniturque adhuc
barbaras apud gentes*."
It is not strange that this notice of a chief, cut off in the
flower of his age by the treachery of those closest to him, has
reminded modern critics of the story of Sigurd the Volsung,
and that some have seen in the story of Sigurd's death by the
wiles of his sworn brethren the last traditional echo of this
treacherous onslaught upon Arminius, and of the songs in
which his deeds and his death were celebrated. No evidence
however is forthcoming to support such a theory, and there
is much which tells directly against it.
1 Germania, 2. 2 Annals, u, 88.
Stories known to Widsith: Gothic heroes 13
Not till some three centuries later, and among the Goths,
do we find the first German heroes whose stories told in German
speech have lasted through the ages.
It may even be that the development of epic poetry, as
opposed to the choric poem of a more primitive age, was the
work of Gothic minstrels. But there is no proof of this.
Those critics1 who would see evidence of the preeminence of the
Goths in epic song in the fact that Clovis the Frank requested
the gift of a harper from Theodoric the Goth, as recorded in
the correspondence of Theodoric's minister, Cassiodorus2, have
overlooked the context. For in the preceding letter, a florid
effusion treating of the Lydian and Phrygian measures, the
Sirens, Orpheus, and the Psalms of David, the selection of this
harper is deputed to Boethius the Patrician. The citharoedus
was, then, no Gothic minstrel, but an Italian or Greek per
former ; and the letter to Clovis, so far from marking the first
spread of epic poetry from the Goths to the Franks, proves
nothing at all3. It is necessary to emphasize this, because
theories as to the date of Widsith have been based upon the
argument that the gleeman's visit to Eormanric reflects that
preeminence of the Gothic court in heroic song which is
inferred from the letter to Clovis4.
Eastgota.
But there is no need to exaggerate the part played by the
Goths in the development of Germanic song, for in any case it
is important enough.
Jordanes mentions songs which dealt with the wanderings
of the Goths after they had left their old home on the shores of
the Baltic, and were migrating across the steppes of Russia to
1 e.g. Koegel, Ltg. i, 1, 130. 2 Varice, n, 40, 41.
3 Since writing this I have noticed that it was pointed out nearly thirty
years ago by an Italian scholar (Eajna, Origini, p. 36) and more recently by
an English one (Chambers, The Mediaeval Stage, 1903, p. 34) that the letter
to Boethius indicates that the citharoedus was sent to the Franks to replace
barbaric by civilized music. Yet he is still quoted as evidence for Germanic
heroic song, even by those who do not accept the whole of Koegel's deductions
(e.g. Symons in Pauls Grdr.(2) in, 622; Chadwick in Camb. Hist. i. p. 20).
4 See _Bojunga in P.B.B. xvi, 548.
14 Widsith
the Black Sea1. It is not however till the third century, with
the irruption of the Goths into the Roman Empire, that we
come to a king — Ostrogotha — whose name became part of the
common stock of Germanic story.
Towards the end of Widsith a list is given of chieftains at
the court of Eormanric, king of the Goths. The list is a long
one, and, as we should expect, it includes the names of Goths
of a much earlier date, who have been attracted into the circle
of the later and greater king. Thus, amongst Eormanric's
followers mention is made of —
East-Gotan,
frodne and godne, feeder Unwenes.
Eastgota, far from being a follower of Eormanric, was
apparently his great-great-grandfather. But his fame paled
before that of younger heroes, and his story was forgotten, amid
the exciting events which crowded one upon another during the
centuries following his reign. He is otherwise unremembered
in extant Germanic heroic poetry, but some hint of his story is
given by Cassiodorus (c. 480 — 575), and by Jordanes, whose
De Origine Actibusque Getarum (written between 550 and 552)
has preserved the substance of the lost books of Cassiodorus, on
the History of the Goths.
Enituit enim Ostrogotha patientia wrote Cassiodorus in a
famous passage on the characteristics of the kings of Gothic
story2 : and this patience, followed in the end by hard hitting,
seems to be the essential feature of the story of Ostrogotha, as
told by Jordanes3.
Stirred by the injustice of the Emperor Philip (A.D. 244 —
249) who refuses to pay to the Goths their due subsidies,
1 ad extremam Scythice partem, que Ponto [Pontico] mari vicina eat, pro-
perant ; qitemadmodum et in priscis eorurn carminibus pene storicu [historico]
ritu in commune recolitur. Getica, rv, (28).
2 Cassiodorus, Varice, xi, 1.
3 Whatever may be the etymology of Ostrogotha (see Streitberg in I.F. iv,
300-9), it is not disputed that at a very early period it was understood to mean
' East-Goth.' For Hunuil-Unwen see note to 1. 114. The equation of Eastgota
and Ostrogotha rests then upon the identity of name, and the close resemblance
of the son's name. We might also compare frod and god with Cassiodorus' enituit
patientia and Jordanes' ut erat solidi animi (xvn, 99). The evidence for the
identification seems satisfactory, and there is nothing to be said for the other
suggested interpretations of the name, as Angantyr (Rafn in Antiquites Russes, i,
111), Eckehart (Heinzel, Ostgotische Heldensage, 67) or as "ein dritter sonst
unbekannter Hereling" (Boer, Ermanarich, 67, 79).
Stories known to Widsith: Gothic heroes 15
Ostrogotha, king of the united Gothic nation, Ostrogoths and
Visigoths alike, crosses the Danube, and wastes Mcesia and
Thrace : on the occasion of a second invasion the Goths put the
town of Marcianopolis to ransom, and return with the plunder
to their land. Hereupon the nation of the Gepidae, closely
related to the Goths, is moved to envy. The Gepidae, and their
king Fastida, have been elated by a recent victory, in which
they have almost annihilated the Burgundians; they accordingly
send envoys to Ostrogotha, and demand that he shall either
meet them in battle, or surrender to them a portion of the
Gothic territory. Ostrogotha replies that he abhors such a
war; that they are kinsmen, and should not strive; neverthe
less, that he will not yield his land. A stubbornly fought
contest follows: but the superiority of the Gothic cause and
the Gothic spirit to that of the "dull" Gepidae gives the victory
to Ostrogotha, and the Gepidae flee back to their country.
During the rest of Ostrogotha's reign the Goths enjoy peace,
till, at his death, Cniva succeeds and again invades Roman
territory1.
How much of this is history, and how much saga, it is not
easy to say. But there seems no sufficient reason to doubt
that Ostrogotha is a real person2. His invasion was fruitless,
and his fame was crowded out by that of later and more
popular Gothic heroes, Widigauja, Ermanaric, Theodoric. But
Ostrogotha's name is the first that meets us in the record of
two among the great things of the world — the Gothic invasion
of the Roman Empire, and the Gothic heroic stories which were
to be the delight of Europe for a thousand years.
Ermanaric and Sonhild.
And ic wees mid Eormanrice ealle tyrage.
Very different was the fortune of Ostrogotha's successor in
the fourth generation, King Ermanaric3, who died about the
year 375, and whose fame remained in the memory of German-
1 Jordanes, Getica xvi, xvn, (89-100). 2 See Appendix C, Eastgota.
3 *Airmanareiks according to the spriba} usage of the Gothic Bible,
16 Widsith
speaking people for at least twelve centuries1. It was, however,
as the " colossal type of a ferocious and covetous ruler, raging
against his own kin," a character which he already bears in
Widsith, that Ermanaric was remembered.
The historic fact of Ermanaric's reign is certified by the
contemporary and reliable evidence of Ammianus Marcellinus2,
who records how Ermanaric by his warlike deeds made himself
formidable to his neighbours, how the Huns and Alans broke
in upon his extensive and prosperous empire and ravaged it,
and how Ermanaric, after vain attempts to rally against the
even greater disasters which rumour threatened, sought refuge
in a voluntary death. A death so unprecedented in the annals
of German kings naturally lent itself to romantic explanation
and comment ; and the story, as given by Jordanes almost two
centuries later, may perhaps contain a considerable element
of fiction.
Jordanes tells of the great power attained by Ermanaric
nobilissimuis Amalorum, quern merito nonnulli Alexandra Magno
conparavere maiores, and how he ruled, not only over the
German races and the Esths of the coast, but also over the
Wends and the Sclaves, " tribes who, although now, thanks to
our sins, they are raging round us on every side, were yet at
that time all subject to Ermanaric3."
This empire, stretching from the Black Sea to the Baltic,
seems to have been as loosely organized as it was widely
spread, and it fell to pieces on the advent of the Huns, of
whose diabolical origin and " bestial ferocity " Jordanes gives an
account, tinged with all the gusto of racial antipathy. Whilst
Ermanaric was brooding over the appearance of the Huns, the
faithless race of the Rosomoni, which, amongst others, was at
1 A Low German poem was printed in 1560, which told how " Dirick van
dem Berne [Theodoric] he siilff twolffte " slew " den Koninck van Armentriken,
mit veerde halff hundert man." Reprinted by Goedeke, Hanover, 1851.
A reprint is also given by Abeling in Teutonia, vn, Supplementheft. This very
corrupt version shows some evidence of being ultimately derived from the same
original as the HarriSismdl. (See Symons in Z.f.d.Ph, xxxvin, 145-166.) The
poem keeps the old tradition as to Ermanaric's ferocity. Dirick says,
' De Koeninck van Armentriken de ys uns suluen gram,
He wil vns Heren all twrelue yn den Galgen hengen laen.'
2 xxxi, 3, 1 (recens. v. Gardthausen, Vol. n, Lipsia, 1875).
3 Getica, xxin, (119).
Stories known to Widsith: Gothic heroes 17
that time tributary to him, took occasion to betray him. For
the king, moved with fury, had caused a certain woman of this
race, named Sunilda, to be torn asunder by wild horses on
account of the treacherous revolt of her husband ; her brothers,
Sarus and Ammius, in vengeance for the death of their sister,
attacked Ermanaric and wounded him in the side, so that he
became sick and infirm. Balamber, king of the Huns, took
advantage of this sickness to move his battle array against the
Ostrogoths, from whom the Visigoths were now, owing to some
dispute, separated.
In the midst of these troubles, Ermanaric, overcome by his
wounds, and the incursions of the Huns, died, full of days, at
the age of one hundred and ten years1.
Jordanes' story has been accepted by the historians as
history2; as a more detailed account of events only partially
known to the contemporary Ammianus. The mythologists, on
the other hand, are convinced that it is myth3, a legendary
1 Nam Hermanaricus, rex Gothorum, licet, ut superius retulimus, multarum
gentium extiterat triumpbator, de Hunnorum tamen adventu dam cogitat,
Bosomonorum (Bosomanorum, Bosomorum, Rosimanorum) gens infida, quee
tune inter alias illi famulatum exhibebat, tali eum nanciscitur occasions
decipere. dum enim quandam mulierem Sunilda (Sunielh, Sunihil; Suani-
hildam, Holder) nomine ex gente memorata pro mariti fraudulento discessu rex
furore commotus equis ferocibus inligatam incitatisque cursibus per diversa
divelli prascipisset, fratres ejus Sarus et Ammius (Animus, Aminus, lammius)
germanaa obitum vindicantes, Hermanarici latus ferro petierunt ; quo vulnere
saucius egram vitam corporis imbecillitate contraxit. quam adversam ejus
valitudinem captans Balamber rex Hunnorum in Ostrogotbarum parte movit
procinctum, a quorum societate iam Vesegothse quadam inter se intentione
(contentione, Holder) seiuncti habebantur. inter hffic Hermanaricus tarn
vulneris dolore quam etiam Hunnorum incursionibus non ferens, grandevus
et plenus dierum centesimo decimo anno vitas suse defunctus est.
Jordanes, Getica ed. Mommsen, cap. xxiv (129-30). For the interpretation
of difficult words: gens = nation, not family: famulatum exhibebat=vto.B subject,
of a whole tribe, not referring to the personal service of a retainer : discessus =
revolt, not flight, see Mommsen's Index (iv) and Jiriczek <2) 57-59.
2 Gibbon, ed. Bury, 1897, in, 57, 91-2 ; Hodgkin's Italy, i, 246 (but cf. in, 4).
3 That the story is fictitious has been argued from the name Sunilda, which
is the form found in the best MSS of Jordanes. This, it is said, is the Gothic
*Sonahildi or *Sonihilds, ' the woman who atones ' ; clearly therefore a feigned
name, like the names of the brothers Sarus and Ammius — ' the armed ones.' See
Miillenhoff in Index to Mommsen's Jordanes ; Symons in Pauls Grdr. (2) in, 683 ;
Jiriczek (2) 63 ; Boediger, Die Sage von Ermenrich, 248. Boediger would make
Jordanes' story purely mythical in origin ; a story of the god Irmin-tiu, other
wise Ermanarikaz, which was later, he thinks, connected with the story of the
historic Ermanaric in order to explain that king's mysterious death. But this
is more than doubtful. Boediger lays weight upon the names Sarus and
Ammius, " Solche einfachen Namen deuten immer auf mythische Trager."
But Sarus, so far from being a fictitious name, was borne by a chief in historic
c. 2
18 Widsith
amplification of the fact recorded by the Roman historian.
There is no conclusive evidence in either direction. The story
of Ermanaric's suicide, as told by Ammianus, is strange and
unaccountable, but it rests on the word of a peculiarly truthful
historian : if it is true, then its very strangeness only gives all
the more reason why a saga should have been invented to
account for Ermanaric's death.
Unfortunately, true or not, Jordanes' account leaves much
unsettled. All attempts to identify and explain the Roso-
monorum gens infida have proved inconclusive, though a
stream of conjectures is forthcoming, beginning with the
attempt, uusuccessful though supported by the great names
of Gibbon and of Grimm, to identify them with the Russians1,
down to the more recent interpretations of the word as the
"red-haired" or "red-skinned," a symbol of perfidy3, or the
" doughty8" or " the men of the ice4."
Jordanes' tale, at best, is obscure, and the difficulty has
been rather increased by attempts to force his words so as to
read into them the version of the Swanhild story, which we
know to have been current in Scandinavia three centuries after
times (Hodgkin's Italy i, 733 ; cf. Jordanes, Romana, 321, with Miillenhoff's
note). As to the name Sunilda, the MSB of Jordanes vary greatly. Sunielh,
Sunihil, are found. The oldest record of the name, in Germany, the charter
of 786, gives the ambiguous form Suanailta ; the Icelandic gives Svanhildr. So
that argument from the form of the name seems perilous; and indeed there
are reasons for connecting the first element in the name rather with sunja
(truth) than son (atonement) ; see Boer, Ermanarich, 8. Even if the name does
mean ' the woman who atones ' this would not involve its having been invented
to fit the story, for Sunilda's death does not lead to any atonement, but rather
is the cause of further feud (Panzer, H.i.B. 41, 77).
1 Gibbon, ed. Bury in, 92 ; Grimm, G.D.S. 747-8. Heinzel, Hervararsaga
(W.S.B. cxiv, 102), thinks the Bosomoni were probably Slavs. For a more
recent identification with the Ruotsi of the Baltic see Aspelin (J. B.) La
Rosomonorum gens et le Ruatsi, Helsingfors, 1884 ; and cf. Grotenfelt, Die Sagen
von Hermanarich u. Kullervo in FinniscJi-Ugrische Forschungen, n, 50 (1903).
2 This not very probable suggestion is made by Bugge (A.f.n.F. i, 1-20).
He suggests an original form *Rusmunans, and points out that the treacherous
Sibech, Sifka, who in the later versions of the story fills the place earlier
occupied by the treacherous Bosomoni, and their chief, as the evil genius of
Ermanaric, has red hair. See Thidreks saga, cap. 186.
Bugge (p. 9) assumes in his argument that the deserter was chief of the
Bosomoni ; this however does not necessarily follow from the words of Jordanes,
who states only that his wife belonged to that race. The treachery of the
Bosomoni is the attack by Sarus and Ammius.
3 Koegel, Ltg. i, 1, 148.
4 *Hrusamans, suggested as a possible alternative to Bugge's interpretation
by Grienberger (Z.f.d.A. xxxix, p. 159. The whole article, on the races noted
by Jordanes as subject to Ermanaric, is important).
Stories known to Widsith: Gothic heroes 19
his day1. With this interpretation Sunilda would be the wife
of Ermanaric and pro mariti fraudulento discessu would mean
" on account of her dishonest desertion of her husband ; viz.
Ermanaric." But this rendering, in addition to doing violence
to Jordanes' Latin2, is, from the context, impossible: if Jordanes
had regarded Sunilda as Ermanaric's unfaithful wife, he would
hardly have introduced her without explanation simply as
quandam mulierem Sunilda nomine.
Except in making Ermanaric's victim his wife, this
Scandinavian version keeps fairly near the original story.
Jormunrekk sends his son, and a treacherous councillor
Bikki, to woo on his behalf, and to escort to the Gothic
court, the maiden Swanhild. Bikki first urges the son to
woo the maiden for himself, and then betrays him to his father.
Jormunrekk in wrath causes his son to be hanged, and has
Swanhild bound in the town gate, that she may be trodden to
death by horses. But she is so fair that no horse will trample
on her, till Bikki counsels that her eyes be bound3. So she
dies, but her brothers Hamthir and Sorli, clothed in enchanted
armour given them by Guthrun their mother, attack Jormun
rekk in his hall and smite off his hands and feet, before they
are overwhelmed by numbers and stoned to death.
The Norse poets do not trouble to make the action of
Bikki the traitor intelligible : no motive is given for his be
traying Swanhild and Jormunrekk's son to death4. Saxo
1 Koegel, Ltg. i, 1, 147. This interpretation was earlier suggested by
W. Muller in his Mythologie der deutschen Heldensage, Heilbronn, 1886, 163-4.
2 Jiriczek(2) 58-9.
3 The account given by Snorri differs in some details from that given in
the Vglsunga saga, and chiefly in this, that the murder of Swanhild is not
deliberate, but the result of a sudden impulse. Jormunrekk, as he rides
from hunting in the forest, sees Swanhild dressing her hair; and he and his
men tread her to death under their horses' feet : Sn. Edda, udg. J6nsson,
p. 189.
4 A portion of the story forms the subject of two poems at the end of the
Elder Edda, the Hamftismdl and Gvftrunarhvgt, which have been attributed to
the tenth century (see Banisch, Zur Kritik u. Metrik der Hambismdl, Berlin
dissertation, 1888, 29, 74), and it is told by Snorri early in the thirteenth
century, and again in the Vqlsunga saga somewhat after the middle of that
century. All these versions differ slightly, and their mutual relations are by
no means clear. Still more important, by reason of their early date, are the
allusions in the Ragnars drapa of Bragi the Old. In this the poet describes the
shield which he has received from llagnar Lodbrok. This shield is covered,
2—2
20 Widsith
Grammaticus tells us that Bicco, before he became the servant
of larmericus, was a Livonian prince, whose brothers had been
slain by that king, and who thus wreaks his vengeance by
causing the king to slay his dearest ; but this is clearly a sub
sequent addition to the story, reflecting as it does the racial
quarrels of a later age.
Such was the tale, as it was current in Scandinavian lands
from at least the ninth century throughout the Middle Ages.
But both in England and in continental Germany this portion
of the Ermanaric story seems soon to have died out. In
England the names Swanilda and Sorli occur in charters;
but very late, whilst their form makes it probable that they
are Scandinavian introductions, not witnesses to native English
tradition: Serlo too is found, but this again is a foreign form1.
But the occurrence of the name of Bicca or Becca fixed to
different English localities, Biccanhlew, Biccanpol, Beccanleah,
Beccanford, seems to show that his part in the story was at
one time known in England ; and this is rendered the more
probable by Widsith, which not only numbers Becca among the
retainers of Ermanaric, but mentions the tribe subject to him —
the Baningas.
like that of Achilles, with images ; the figures on Bragi's shield, however, refer
to specific stories, and the allusions show that the Swanhild story was known
in the North, in the first half of the ninth century, in substantially the same
form as we have it in the Volsunga saga, some four and a half centuries later.
The authenticity of Bragi's song, which forms an important landmark in the
history of German heroic tradition, has been disputed by Bugge, on the ground
of grammatical forms, vocabulary, metre, style and subject matter. (Bidrag
til den eldste skalde-digtnings historic, 1894.) These arguments have been
weighed at great length by F. J6nsson (Aarbpger f. nord. Oldk. u, 10, 4, 1895
De celdste skjalde og deres kvad), and more briefly by Gering (Z.f.d.Ph.
xxvni, 111), Kahle (Litter aturbl. f. PhiloL, 1895, Sept.), Mogk (Lit.-Centralblatt,
1895, 15, 540), Panzer (H.G. 155-7), Jiriczek (384), with the unanimous verdict
that the case against the authenticity of the Eagnars drdpa has not been
made out.
The version of the story given by Saxo Grammaticus may be regarded
as a compromise between the Norse and the (mainly lost) Low German versions.
The different elements have been separated by Olrik, Sakses Oldhistorie, n, 1894,
252-3 ; in the imperfect state of our evidence it is hardly possible to say
which element predominates. For different views on this subject see Symons
in Pauls Grdr.fr) in, 688; Jiriczek(2) 115. Ermanaric's son is called Bandve'r
in the Norse versions, Broderus in Saxo.
1 Binz in P.B.B. xx, 209 thinks it Frankish. If the references in the
Quedlinburg and Wurzburg chronicles to the death of Ermanaric are really
derived, as Schroeder has argued (Z.f.d.A. XLI, 24-32) from an English source,
then the story of Sarus and Ammius must have been still known in England
in the ninth century. But see below, p. 30 (footnote).
Stories known to Widsith: Gothic heroes 21
Eormanric and Ealhhild in Widsith.
That Ermanaric was well known in England is proved by
the references in Old English heroic poetry.
In Beowulf there is mention of the "wily hatred1" of
Eormenric. Not dissimilar is the reference in Dear's Lament:
We have heard of Eormanric's wolfish mind : far and wide he ruled
the folk of the kingdom of the Goths : a grim king was he.
But the fullest and clearest reference to Eormanric is in the
central passage of our poem :
And I was with Eormanric all the time, there the king of the Goths
was bounteous unto me. Lord of cities and their folk, he gave me an
armlet, of pure gold, worth six hundred shillings. This I gave into the
possession of my lord and protector Eadgils, when I came home : a gift
unto my beloved prince, because he, lord of the Myrgings, gave me my
land, the home of my father. A second ring then Ealhhild gave unto me,
noble queen of chivalry, daughter of Eadwine. Through many lands her
praise extended, when I must tell in song, where under the heavens I
best knew a queen adorned with gold giving forth treasure ; when Scilling
and I with clear voice raised the song before our noble lord (loud to the
harp the words made melody) what time many cunning men said they had
never heard a better song. Thence I wandered through all the land of
the Goths...
It is remarkable that whilst certain points in the criticism
of Widsith have been scrutinized and canvassed with almost
excessive keenness, others have been taken on trust from the
days of the first editors. Since Ealhhild is here mentioned im
mediately after Eadgils, and since Widsith the Myrging owes
obligations to both, Leo conjectured that Ealhhild was queen of
the Myrgings and wife of Eadgils2. In this he was followed by
Lappenberg8. Both brought forward this relationship as a
mere inference. Later scholars4 have followed them, but often
without any expression of doubt. Yet there is nothing in the
poem which points at all conclusively to this relationship. All
we know about Ealhhild is that she is a daughter of Eadwine,
who is apparently the same Eadwine, king of the Lombards,
who is mentioned as ^Elfwine's father.
1 searonfyas. The word, however, need not necessarily have a bad sense,
cf. Beowulf, 582.
2 durch Ealhhilden...die (wie es scheint) Fiirstin der Myrgingen (wohl
Eadgils Gemahlin), p. 75.
3 Ealhhilde wird erwahnt, wie es scheint, als Fiirstin der Myrgingen, Gemahlin
oder Mutter des Koniges derselben Eadgils, p. 175.
4 e.g. Thorpe (510), Ettmullerd) 12, Miillenhoff, Beovulf, 98 ; Ten Brink in
Pauls Grdr. (j n, 1, 544 ; Koegel, Ltg. i, 1, 139 ; Stopford Brooke, i, 2.
22 Widsith
It was left for Richard Heinzel, who was not dealing
primarily with our poem at all, and who consequently ap
proached it without any prepossessions, to suggest that
Ealhhild is the wife, not of Eadgils, but of Eormanric : and
that we have a reference to a form of the story in which
Ealhhild had been substituted for Swanhild, and the Lombard
dynasty for the " treacherous Rosomoni." Heinzel threw out
this suggestion cursorily, without discussing it at length, and it
has not yet been sufficiently scrutinized1. Yet the interpreta
tion of Widsith depends upon it. It is remarkable how many of
the theories of previous critics have been built upon the as
sumption that Ealhhild is the wife of Eadgils.
The discrepancy in the names Swanhild and Ealhhild ought
not to weigh too heavily against Heiuzel's view. Alike in epic
song and in everyday conversation, Swanhild would be ab
breviated to Hild. In song, an abbreviated Hild might easily
be expanded to Ealhhild. Both changes can be exactly
paralleled. Thus, in the saga of Olaf Tryggvason, Jarl
Eystein has a daughter named Hild or Swanhild2. The
change by which Hild(ico) may have become Grimhild
(Kriemhild) is a commonplace of the history of the develop
ment of the Niblung story8. So that there would be nothing
unprecedented in Swanhild being first abbreviated to Hild and
then expanded to Ealhhild4. But we have seen that it is quite
uncertain what the first element of the name originally was.
1 Ueber die Hervararsaga, p. 102 in the W.S.B. cxiv, 11. Heinzel's view
has been accepted by Jiriczek(2) 73, Lawrence in Mod. Philol. iv, 351-3,
Holthausen, n, 164, and, with some doubt, by Gummere(2) 199; it has been
developed by Boer (Ermanarich, 15-18).
2 " M6'Sir J>eirra var Hildr e$r Svanhildr, dottir Eysteins jarls," Fornmanna
Sogur, 1825, i, 5. For abbreviations of names see Franz Stark, Die Kosenamen
der Germanen, Wien, 1868; and a note by Olsen in the Aarbjger for Nordisk
Oldkyndighed, 1905, p. 72.
3 See Symons, Sigfrid u. Brunhild in Zachers Z.f.d.Ph. xxiv, 29 ; and, to
the contrary, Boer in Z.f.d.Ph. xxxvn, 484. Of., too, Abeling, 206.
4 Although compound names in Ealh- are not uncommon in Old English,
the name Ealhhild does not seem to have been current in England in historic
times. It is found on the continent, though rarely. In the ninth century
revenue book of the Abbey of St Eemi at Reims the name occurs twice
(Polyptique de VAbbaye de Saint Remi, par B. E. Guerard, Paris, 1853.
Alchildis ingenua, p. 69 ; Algildis ingenua, cum infantibus, p. 86). In 819
a serf named Alachilt is granted, with many others, to the Abbey of Fulda
(Codex Diplomaticus Fuldensis, herausg. Dronke, Cassel, 1850). The name
Alachilt, Alahilt occurs also in the Books of the Confraternity of St Gall
(ed. Piper in M.G.H. Berlin, 1884). See Forstemann, i, pp. 53, 75.
Stories known to Widsith: Gothic heroes 23
Probably it was Son- or Sun-, and only later, when it became
unintelligible, was this corrupted into Swan. Hild is the
fixed element in the name ; and there is no reason why in some
versions of the story Sunhild should not have been altered to
Ealhhild. We constantly find the compound names of heroes
thus changed in one of their elements. The Hrothgar of
Beowulf is the Roe, Hroar of Scandinavian tradition, forms
which would correspond to an Old English Hrothhere. So
the Eanmund of Beowulf appears in Saxo as Homothus
(i.e. Eymothr instead of Eymundr)1. The Sifrit of the South
is the Sigurd of the North. If even so great a hero as
Sigfrid could not keep his name intact in passing from one end
of Germania to the other, there is no inherent improbability
in the name Sonhild — Sunhild — Swanhild having been cor
rupted.
It becomes then a question of what evidence we have in
favour of identifying the Ealhhild of our poem with Swanhild.
At least five points, some of them not strong in themselves,
deserve to be weighed :
(1) Ealhhild is daughter of Eadwine, apparently the famous
Lombard king who is mentioned elsewhere (1. 74) in Widsith ;
and quite apart from Heinzel's theory, it is thought that in
Widsith the Lombard story has, in some way or other, been
linked on to the Gothic. How this took place we cannot say,
but that Lombard heroes are enumerated among Eormanric's
household was pointed out half a century ago by Mullenhoff2,
and seems quite possible. Now this contamination of two dis
tinct traditions would be explicable if the English version of
the story gave Eormanric a Lombard princess to wife. And
this might easily happen. The Rosomoni, to whom Sunhild
originally belonged, had long been forgotten. Yet Eormanric's
wife must belong to some great race. To whom rather than to the
Lombards, who had just established themselves as successors
of the Goths in the lands of Italy ? There was more reason to
turn Eormanric's luckless queen into a Lombard than into a
1 Cf. Olrik, Sakses Oldhistorie, 191. For other changes in names of heroes
of tradition see Heusler in Z.f.d.A. LII, 97, etc. : Heldennamen in mehrfacher
Lautgestalt.
2 Z,f.d.A. xi, 278.
24 Widsith
Hellespontine, which she is in Saxo1, or a Volsung, which she is
in the VQlsunga saga.
Yet we must not place too much weight upon this
argument: for it is not absolutely certain that our poet re
garded Ealhhild as a Lombard princess, whilst the supposed list
of Lombard heroes inserted into the catalogue of Eormanric's
household is far too doubtful for us safely to argue from it.
The supposed combination of Gothic with Lombard legend
does not, then, go very far towards proving Heinzel's
hypothesis.
(2) This hypothesis makes it easier to explain why our
poet, whilst he makes Eormanric give to Widsith a ring of
very high price, yet calls him a wraty wcerloga. Remembering
how highly generosity to retainers in general, and to singers in
particular, ranks among the kingly virtues in Old English heroic
poetry, it is likely that some specific evil act lies in the back
ground in order to justify this epithet in this context.
The Ermanaric of tradition certainly was a "treacherous
tyrant " : but it seems hardly fair that he should be called so in
the poem celebrating that Widsith whom he had treated so well.
But if Widsith owed obligations to Ermanaric's murdered wife,
then the epithet becomes appropriate enough.
(3) A detailed examination of the passage dealing with
Eormanric's gift to Widsith favours the interpretation that
Ealhhild is the queen of Eormanric, who, after the bountiful
act of her lord, gives Widsith a second ring in the Gothic hall,
rather than the queen of Eadgils, who gives him the ring on
his return home, in compensation for that which he presents
to her husband. The episode deserves consideration for the
light it throws on the manners and ideals of ancient Germania.
When trade, or fortune in the gift hall or on the battlefield,
had put into the hands of a " thegn " treasures too great for a
private gentleman to possess — an armlet or necklace of sur
passing value, a king's panoply, a tame bear — custom approved
of his presenting it to a great chief: if he had one, to his own
chief. But it was not usual for the chief to acknowledge this
gift by presenting in return things like in kind but inferior in
1 Ed. Holder, vni, 279-81.
Stories known to Widsith: Gothic heroes 25
quality. The prince makes the donor his personal servant, or
gives him a landed estate, or weds him to his daughter. It is
not so much an exchange as a recognition of mutual dependence.
The winnings of the retainer fall to the chief: the chief values
wealth only as he may share it with his retainers1. Thus
Beowulf receives from Hrothgar rich gifts for his service at
Heorot, sword, armour, horses2 ; jewels, and a glorious necklace
from Hrothgar's queen3: on his return he gives armour, sword
and horses to his lord Hygelac, and the necklace to Hygd,
Hygelac's wife. On his part, Hygelac gives Beowulf feudal
domains, placing, as he does so, in Beowulf s bosom the sword
of their common grandfather Hrethel4. Again, when Eofor the
Geat slays the Swedish king Ongentheow, he presents the spoils
to Hygelac, who gives him in return land and wealth, and the
hand of his only daughter5.
A not dissimilar episode comes in the Saga of Harold
Hardrada6. Authun, an Icelander, comes to the court of
Harold with a bear which the king covets. However, Authun
will neither sell nor give the beast to the king, but declares
his intention of taking it to Swein of Denmark, with whom
King Harold is at war. Harold allows him free passage ; but
on his way back he is to come and tell what King Swein has
given in return. This Authun does,
The king took his greeting well. "Sit down," he said, "and drink
here with us," and Authun did so. Then Harold asked, " How did King
Swein requite thee for the bear ? " Authun replied, " In that he accepted it,
Sir, from me." " I would have requited thee so," said the king, " what
more reward did he give thee ? " Authun replied, " He gave me silver to go
South " (on a pilgrimage to Rome). Then said King Harold, " To many
men does King Swein give silver, though they have not brought him
treasures. What more?" "He offered," said Authun, "to make me his
page and do me much honour." " That was a good offer," said the king,
"but he surely did more." " He gave me," said Authun, " a big ship with
a cargo of the best." " That was magnificently done," said the king, " but
so would I have requited thee. Did he give thee more 1 " Authun said,
" He gave me a leather bag full of silver, and said that then I should not
be destitute, if I kept that, though my ship were wrecked off Iceland."
"That was an excellent precaution to take," said the king, "and more
than I should have done. I should have thought we were quits when I
had given thee the ship and cargo : did he requite thee yet more ? " " Of
1 Cf. Tacitus, Germania xiv. 2 1020-49.
3 1192, etc. * 2152-2199. 8 2990.
6 Fornmanna Sogur, Kaupmannahtffn, 1831, vr, 297-307: Wimmer, Oldnordisk
Laxelog, 1877, 55-60. The two texts differ in detail.
26 Widsith
a truth, Sir," said Authun, " he did : he gave me this ring which I have on
my hand, and said it might happen that I lost all my wealth and the
ship : that then I should not be destitute if I had the ring. And he bade
me not part with it, unless I owed so much to some great man that I
wished to give it him. And now I have met such a man : for thou
mightest have taken from me both the bear and my life : but thou didst
give me free passage when others obtained it not." The king received the
gift graciously and gave Authun in return good gifts before they parted.
But if Queen Ellisif had stepped down from the dais and
offered Authun an inferior ring in exchange for the gift he had
just made to King Harold — Authun would no doubt have ac
cepted it with his usual tact : but I think every retainer in the
hall would have felt that a mistake had been made. Nor would
Queen Hygd have given Beowulf another necklace in exchange
for the glorious one which he gave her.
Now the armlet given by Eormanric is of extraordinary
value : for Eormanric was the richest and the most generous
of tyrants. It is valued at six hundred shillings1. It would
probably have been held ostentatious, and it would certainly
have been imprudent, for a minstrel to wear treasure worth
the price of his own life. Nobody but Eormanric, the possessor
of fabulous wealth2, could afford to give such an armlet to
a singer.
In giving this present to his lord, the reward Widsith would
expect would be, as Authun said, "that he should accept it."
He certainly would not expect an inferior ring in return. It is
turning a gracious act of love and loyalty into a matter of mere
barter if we are to read the lines as Ten Brink and the rest
read them :
Eadgils' Gemahlin aber, Ealhhild, entschddigte den Sanger fur das
verschenkte Kleinod durch einen andern Ring 3.
On the other hand, if Ealhhild is the Gothic queen, her
action is in accordance with the laws of Germanic courtesy. She
follows up the magnificent present of Ermanaric, her lord, by
offering the poet a second ring : one of less value, naturally ;
but which he is to keep for her sake, in memory of the service
1 See Appendix N, On the beag given by Eormanric.
2 See Saxo, Book VIH (ed. Holder, p. 278), for an account of the treasure
house of larmericus : cf. too Dietrichs Flucht, 7854-7862.
3 Ten Brink in Pauls Grdr.^) n, 542. So Lawrence speaks of it as a
"business transaction " (Mod. Philol. iv, 369).
Stories known to Widsith: Gothic heroes 27
he has done her in escorting her to the hall of the Goth. And
this ring Widsith does keep.
An argument which postulates gentlemanly feeling among
sixth century barbarians may seem to some fantastic. But it
can be further urged :
(4) That only by this interpretation does the whole
passage yield consistent sense. " Eormanric gave me a ring,
which, when I came home, I gave to Eadgils my lord (minum
hleodryhtne). Ealhhild gave me another ring. Whenever I
had to sing, I sang her praises through many lands as the most
generous queen : when Scilling and I sang before our lord
(uncrum sigedryhtne), many men said they never heard a better
song. Thence I fared through the whole land of the Goths."
If the poet rewards his benefactress by spreading her fame
through many lands, then Eadgils' land, which is instanced, can
hardly have been hers1. Nor can we understand the transition
to the next sentence, " Thence I fared through all the land of
the Goths," if Ealhhild's gift takes place, not in the heart of the
Gothic land, but after Widsith has returned home to the far-
distant Myrging court2.
How, if we make Ealhhild the wife of Eadgils, are we to
interpret the second reference to Widsith's lord (uncrum
sigedryhtne) ? It is inapplicable to Eormanric ; for Widsith is
giving an example of how he rewarded his patroness for her
gift by spreading the fame of her generosity. He cannot sing,
in the hall of Eormanric, of a generosity which he does not
experience till he returns home to Eadgils' hall. Besides, the
word dryhten in Old English poetry always implies a distinct
tie of subjection. In the mouth of Widsith dryhten should
refer to his feudal chief Eadgils, not to Eormanric. But this
it can hardly do, if Ealhhild is the wife of Eadgils. For as we
have seen, Widsith is giving an example of how he spread her
fame through many lands, and he cannot have done this by
singing, in her presence, in her lord's hall.
(5) Finally, and this is the heavy evidence, the introduction
1 This struck Ten Brink. The transition, he says, is abrupt (p. 543).
2 This difficulty had before been felt by Moller (V.E. 2, 3), who, assuming
Ealhhild to be Eadgils' wife, found it necessary to rearrange the text.
28 Widsith
tells us that Ealhhild accompanied Widsith to the house
(not of Eadgils the Myrging, but) of Eormanric the Goth :
and for what purpose except to become his bride, is not
clear1. It has been suggested that she went as a hostage2.
But this does not fit in with the theory of those who hold
that she is wife of Eadgils ; for, if Widsith escorted her as a
hostage to the court of Eormanric, how does she come to be in
Eadgils' hall, when he returns home, to present him with
another armlet? And, though we know from Tacitus that
the daughters of great houses8 were often sent as hostages,
we may well doubt whether a self-respecting prince, such as
we may suppose Eadgils to be, would have sent his wife.
True, many critics have regarded the introduction as a later
addition ; and almost certainly they have been right in so
regarding it4. But in any case the introduction dates from
a time when Old English verse and legend were living; the
poet who prefixed it had our poem before him in a form
certainly much purer and possibly much fuller than that
in which it has come down to us; and the fact that he
evidently regarded Ealhhild as wife of Eormanric is strong
evidence that the original poem implied that she was so.
For these reasons it seems best to regard Ealhhild as the
murdered wife of Eormanric, the Anglian equivalent of the
Gothic Sunilda and the Northern Swanhild. Perhaps however
we ought to discriminate. That Ealhhild is Eormanric's wife
appears to me to be demonstrable : that legend attributed to
her the same tragic story as to Swanhild is probable enough,
but not equally demonstrable.
Ermanaric and the Harlungs.
fferelingas,
Emercan sohte ic and Fridlan.
That the story of Ermanaric and Swanhild was still current
in the eighth century in continental Germany is rendered
probable by the occurrence of the name Suanailta, in con
nection with other names of the Ermanaric cycle, in a charter
1 See note to 1. 6. a Chadwick (138).
8 puellje nobiles. Ger. vin. 4 See below : Chap, iv, end.
Stories known to Widsith: Gothic heroes 29
of the year 786. Unless a very striking series of coincidences
has taken place, the story must have been in the minds of the
god-parents who gave the names1. Yet, alike in High and Low
German tradition, Swanhild was soon forgotten, and with her
the motive for the death of Ermanaric's son2. But another evil
deed was attributed to Ermanaric : the murder of his two
nephews, the young Harlungs, brought about by the evil
counsel, perhaps in the earliest stories, certainly in the later
ones, of his henchman Sibeche.
From the tenth century onwards there are several references
to this Ermanaric story in the Latin chronicles of Germany: of
these the most important is one which is found, with only slight
variations, in the chronicles of Quedlinburg and of Wiirzburg3,
and which introduces the names of three important actors in
the story who are known to us from Widsith, Freotheric and
the Harlung brethren (Herelingas) Emerca and Fridla. These
two chronicles date, the first a few years before, the second
a few years after the year 1000 — and the story they tell is
this :
At that time Ermanaric ruled over all the Goths, more cunning than
all in guile, more generous in gifts ; who after the death of Frederic his
only son, brought about by his own wish, hanged from the gallows his
nephews, Embrica and Fritla. He drove likewise from Verona Theodoric
his nephew at the instigation of his nephew Odoacer, and compelled him
to dwell with Attila in exile... Ermanaric, King of the Goths, was slain in
shameful wise, as was fit, by the brethren Hamido, Sarilo and Odoacer,
whose father he had killed, after they had first cut off his hands and his
feet*.
1 See Miillenhoff in Z.f.d.A. xn, 302-6. On the other hand, it might be
argued that, without postulating anything more than the ordinary working of
the laws of chance, an occasional combination recalling the grouping of heroic
Btory might be expected, in view of the fact that the records of proper names
which have come down to us from the ' Dark Ages ' are comparatively numerous.
2 Yet different forms of the names Sonhild and Swanhild continue in frequent
use. See Pertz passim, especially SS. t. vi; and Langebek iv, 83.
8 Chronicon Wirziburgense in Pertz, fol. SS. t. vi, 1844, p. 23.
4 Eo tempore Ermenricus super omnes Gothos regnavit, astutior omnibus in
dolo, largior in dono ; qui post mortem Friderici filii sui unici, sua perpetratam
voluntate, patrueles suos Embricam et Frithlam patibulo suspendit. Theodericum
timilitfr patruelem suum, instimulante Odoacro patruele suo, de Verona pulsum
apud Attilam exulare coegit. ...Ermenricus rex Gothorum a fratribus Hamido et
Sarilo et Odoacro, quorum patrem interfecerat, amputatis manibus et pedibus,
turpiter, uti dignus erat, occisus est.
We cannot be certain whether this exactly represents a current version, or
is the chronicler's attempt at harmonizing different stories known to him.
The Wiirzburg Chronicle is quoted in preference to the Quedlinburg Annals,
because it tells the story more clearly than does the extant text of Quedlinburg;
30 Widsith
This gives a very rough summary of the story, as current in
continental Germany during the Middle Ages. Though Swan-
hild has gone, it is still remembered that Ermanaric, intentionally
or unintentionally, caused the death of his son, or sons1. But
it is more as the betrayer of his nephews, the " Harlungen," and
still more as the foe of his nephew Theodoric or Dietrich, that
Ermanaric is remembered.
In one of the papers which Mullenhoff left behind him,
unpublished and indeed hardly legible, the origin of the
Harlung story, in the light of comparative mythology, is dis
cussed with a marvellous erudition. The paper, deciphered by
the pious care of a disciple, was published in 1886 2: it is one of
the most valuable of MullenhofFs many contributions to the
history of German heroic song. According to Mullen hoff the
Harlungen were originally twin divinities, comparable to the
Castor and Pollux of Grecian story. They are sent by the
Heaven-God, Irmin-tiu, the Germanic equivalent of Zeus or
Jupiter3, to woo for the God the sun-maiden, bedecked with
although it is quite possible that the entry in Wilrzburg is derived from an
earlier MS of Quedlinburg. (See H. Bresslau, Die Quellen des Chronicon Wirzi-
burgense in the Neues Archiv der Gesell. f. altere deut. Geschichtskunde, xxv,
1900, pp. 32-3.)
Schroeder has argued (Z.f.d.A. XLI, 27, etc.), from the forms of the proper
names, that this reference to the Ermanaric story was derived from an English
source. I cannot agree with this view. The form Adaccaro, found in Quedlinburg
for Odoacro, is hardly English. Schroeder argues that the use of the contracted
form Frithlam points to England. But we are not certain that the original
form in the source of Quedlinburg and Wiirzburg was contracted : the Bamberg
Codex of Wilrzburg gives Frithelam (see Bresslau). And in any case it is not
clear that the contracted form was peculiarly English : against Fridla in Widsith
we have Frfyela in the charter (Birch, 1002 : see below, p. 34).
1 Cf. Dietrichs Flucht 2457, etc.
Ez gewan der KUnic Ermrich
einen sun, der hiez Friderich,
den er sit versande
hin ze der Wilzen lande.
dar an man sin untriuwe sach:
nu seht wie er sin triuwe brach
an sinem liebem kinde!
Cf. also Thidreks saga c. 278-80, 303, 307.
2 Z.f.d.A. xxx, 217- Frija und der Halsbandmythus.
8 That Tin was equivalent to Zeus, phonetically and mythologically, was,
when Mullenhoff wrote, regarded as one of the few pieces of firm land in the
mythological quagmire (cf. Noreen, Altisldndische Grammatik, Halle, 1892,
§ 68, 7). That the two names do not exactly correspond phonetically has been
pointed out, however, by Streitberg (I.F. i, 514). But Streitberg cites parallels
for the phonetic variation, and adheres to the identification on mythological
Stories known to Widsith: Gothic heroes 31
jewels, chief among which is the " Brisingo meni." They are,
however, false to their trust, woo the maiden for themselves,
and fall victims to the wrath of the god whom they have
deceived. Mullenhoff argued that the likeness of this story
of the vengeance of Irmin-tiu upon the Harlungs to the story of
the vengeance of Ermanaric upon his wife and son, and the
similarity of the name of the avenger in each case, caused the
two to be confused : and the Gothic king came to be credited
with having slain, not only his son, but also his nephews, the
two young Harlungs.
The glamour which must always surround the posthumous
work of a great scholar, and the learning and skill displayed
by Mullenhoff in piecing together his materials, disarmed
criticism for the moment ; and in the best recent surveys of
Germanic heroic story and mythology Miillenhoff's mosaic of
myth has been accepted1. Yet it is pure conjecture2. We
have no proof that the " Brisingo meni " was ever in the hands
of the Harlungs, or that its possession by Ermanaric is in any
way connected with their death. Indeed it is quite conceivable
that the Harlungs are derived not from any mythical figures,
but from some historic tribe or princely house, perhaps actually
overthrown by Ermanaric, perhaps supposed in later story to
have been so overthrown3.
t
Now whether Ermanaric was from the beginning the typical
tyrant and oppressor of heroic legend is not clear. We cannot
be certain whether his evil fame is due to the attribution to
him of the murder of the Harlungs4, or whether this attribution
is due to his evil fame5. But even the blackest of tyrants needs
grounds. Bremer (I.F. in, 301-2) regards the phonetic difficulties as in
superable, and points out that the German Tiwaz is the war-god, and not the
Heaven-Father. Yet the identification seems fairly certain. Cf. Mogk in Pauls
Gfrdr.(2) in, 313; B. Meyer, 178; Golther, 200, etc.
1 Symons in Grdr.<2) in, 685; Jiriczek(i) 100; B. Meyer, 218-9 (broadly).
2 Cf. Panzer, H.i.B. 84, etc. 89, etc. ; Boer, Ermanarich, 66.
3 Cf. Boer, Ermanarich, 80. The Harlungs cannot be derived from the
Heruli, as suggested by Grimm (G.D.S. 472) and recently urged by Matthaei
(Z.f.d.A. XLIII, 319, etc.). Harlung and Erl are too different linguistically to
be easily identified, and there is no evidence demanding such identification.
See note to 1. 87.
4 Jiriczek(?) 66, 103; Bojunga in P.B.B. xvi, 548.
8 Symons in Pauls Grdr.(a) m, 685.
32 Widsith
some motive for his evil acts, and this motive was given by the
evil counsellor Sibich [Sifka in the Thidreks saga] who, together
with his antagonist and foil Eckehart, the typical faithful
retainer, may have belonged to the Harlung story even from
the beginning1. Sibich, like lago, does evil for the love of it :
but his actions in the North German story, as recorded in the
Thidreks saga, are made more plausible by the motive of
revenge : his wife has been seduced during his absence from
the country by Erminrek : and he sets to work to bring about,
by his evil counsels, the downfall of the tyrant. By the advice
of Sifka and of Sifka's wife, Erminrek causes the death of his
sons; of two unintentionally, and of the third deliberately2; he
then compasses the death of his two nephews from Harlung
land3, and finally is persuaded to attack his nephew Thithrek
— Theodoric of Verona.
There is an obvious parallel between the original Harlung
myth of the faithless wooers, as it has been reconstructed by
Mtillenhoff, and the Norse story of the prince who, sent to woo
Swanhild, betrays his trust. It is quite possible that Norse
tradition, though it does not record the Harlung myth4, derives
from a version which has nevertheless been influenced by it,
and that the modifications of the story which make Swanhild
the wife of the tyrant, and his own son the object of that
tyrant's jealousy, are due to the influence of the Harlung
story. The difficulty of this supposition is that it leaves us
nothing but the likeness of the names Irmin-tiu and Ermanaric
to account for the contamination of the two stories5. Should
the jealousy motive of the Norse story be simply a reflection of
the Harlung myth it might follow that Bikki is merely a
1 This however is doubtful, especially in view of the absence of Ecgheard
from Widsith. See Binz in P.B.B. xx, 209-10; Panzer, H.i.B. 85; Boer,
Ermanarich, 72, etc.
2 Erminrek has in this version three sons, Frithrek, Eeginballd, and
Samson (Thidreks saga, 277-80).
8 of Aurlungalandi (Thidreks saga, 281).
4 Saxo's reference to the Harlung story is due, no doubt, to Low German
influence.
6 See, however, Boediger in Weinholds Zeitschrift i, 247. Koediger's
suggestion that Ermanarikaz was an old title of Tiwaz would make the
contamination of his story with that of the historic king more intelligible. But
any evidence that this title ever belonged to Tiwaz is wanting.
Stories known to Widsith: Gothic heroes 33
duplicate, a reproduction, of the faithless Sibich1. On the
other hand, it is more likely that they are independent
figures, Bikki the traitor proper to the Swanhild story, Sibich
to the Harlung story2; and this is supported by the mention in
our poem of both figures, Becca and Sifeca.
Eormanric and the "Herelingas" in Widsith.
The mention of the Harlungs in Widsith is important, as
showing that already their story formed a part of the Ermanaric
cycle. In this connection it may be noted that, in Beowulf, the
Brosinga mene is in the hands of Eormenric, and is carried off
by Hama. Whether there was any relation between this and
the Harlung story we cannot tell.
From the absence of further references in Old English
literature, and of persons named after the characters of the
story, Eormanric, Emerca, and Fridla, it has been argued that
this story of Eormanric died out early in England. A Kentish
King lurmenric3 is recorded; elsewhere the name does not
meet us. But it may well have been that these names were
regarded as too ill-omened. We need to be careful in deducing
arguments as to survival of legend from personal names.
That the Harlung story had at one time a footing in
England is proved by the occurrence of Herling, principally in
place names4. We cannot draw conclusions from the fact that
we find no Englishman with the name Sifeca or Seofeca :
.perhaps his story was known too well for any father to give
the traitor's name to his son. It is noteworthy that in the
topography of Saxon Berkshire we find the name of the traitor
not far from that of one of his victims, Seofocan wyr® near
1 Jiriczek(2) 112.
2 Symons in Pauls Grdr. (2) in, 686. In the earlier edition of the Grundriss
Symons had hazarded the guess that Bikka was the unrecorded name of the
husband of Swanhild, Grdr.^) n, 41. In this he had been partly anticipated
by Bugge (Arkiv f. Nordisk Filologi, i, 8, 1883), "Der synes...at vsere god Grund
tU at gjenfinde denne troLtfse Baadgiver i Sunildas af Jordanes omtalte Mand."
3 He was father of Ethelberht of Kent (see the early ninth century
genealogies in Sweet's Oldest Eng. Texts, 171, and cf. Florence of Worcester,
ed. Thorpe, 1848, i, 12).
4 Binz in P.B.B. xx, 209.
c. 3
34 Widsith
Frilpela byrtg1. There is then no reason to doubt that the
Sifeca mentioned in Widsith is the traitor of the Harlung
story2.
Do these two stories of Swanhild and of the Harlungs
involve a different view of Erraanaric : and, if so, can we trace
in Widsith different strata, different views of the character of
the great Gothic tyrant ?
It has been asserted with some confidence that we can3: that
in the oldest Gothic story Ermanaric was a national hero, who
only slowly, and especially through the murder of the Harlungs —
a story which had originally, as these critics suppose, nothing to
do with him — passed into the typical tyrant, the murderer of
his kin. Thus it has been argued by Bojunga that :
" This darkening of Ermanaric's character has not yet taken
place in the older parts of Widsith, although the Harlung
story appears to be already known. In v. 88 ff. Ermanaric
appears as a generous patron of the arts, whilst in the late
introduction (v. 8 — 9) he is, in accordance with later story,
characterized as wrd}> wcerloga."
Here again, the deductions hardly seem warranted by the
facts. There is nothing inconsistent in a tyrant being a
generous patron of the arts : and Widsith is not alone in
asserting the open-handedness of Ermanaric4. It was his
essential feature in story to be at one and the same time
" more cunning than all in guile, more generous in gifts5."
1 Miillenhoff in Z.f.d.A. xxx, 225 ; Birch, Cart. Sax. in, 201 (No. 1002).
2 This has been disputed by Binz, P.B.B. xx, 207-8. Heinzel has shown,
conclusively, that two of the champions mentioned as retainers of Eormenric
are characters of the Hervarar saga. Binz would go further, and identify
Sifeca with Sifka, the Hunnish mistress of the Gothic king Heithrek of the
Hervarar saga. But phonetically this has no advantage over the old identifi
cation of Sifeca with the traitor, whilst in other respects it is inferior : for the
name occurs in Widsith as that of a retainer of Ermanaric, which Sifeca the
traitor was, but Sifka the Hunnish concubine certainly was not. The conjecture
of Binz has, however, been supported by Jiriczek(2) 73; Panzer, H.i.B. 84;
Boer, Ermanarich, 60 (with some hesitation).
3 Bojunga in P.B.B. xvi, 548. Similarly Holier has seen between 11. 8, 9
and 89 a discrepancy pointing to difference of authorship ( V.E. 33). Thorpe
in his edition tried to account for the supposed inconsistency by assuming
a gap in the text. See note to 1. 9 of text.
4 Compare, for example, Alpharts Tod, 207: Thidreks saga, 322.
5 Astutior omnibus in dolo, largior in dono. Chronicon Wirziburgense in
Pertz, SS. t. vi, 1844, p. 23.
Stories Imown to Widsith: Gothic heroes 35
Nor can we be at all certain that any such distinction as
Bojunga assumes, differentiated the Ermanaric of sixth century
tradition from the Ermanaric of, say, eighth century tradition.
No doubt as time went on Ermanaric became more and
more the typical tyrant, and was made responsible for one evil
deed after another. The murder of his wife and the treachery
towards his nephews are clearly not historical : we can trace the
different accretions of tradition by which these crimes come to be
attributed to the great Gothic king. But already in the sixth
century Ermanaric had probably the reputation of a cruel
tyrant. For Cassiodorus, in summing up the characteristic
virtues of all the Amal kings, as mirrors in which the princess
Amalasuintha may perceive her own merits, makes no mention
of him1. Nor is it quite the case, as has been urged, that this
omission can be accounted for by the fact that there was no
tertiwm comparationis between the victorious Ermanaric and
the daughter of Theodoric ; no virtue belonging to Ermanaric
which could be attributed to a woman2. His open-handedness
might certainly have been quoted, the more so as largitas, the
most essential of royal virtues to the Germanic mind, has no
representative in Cassiodorus' list.
The omission, then, of the greatest name in the whole list
of Amalasuintha's ancestors can only be the result, either of a
fault in our text3, or of a feeling that the great conqueror was,
after all, not quite a credit to the family. If then Ermanaric
left behind him a doubtful fame, even among those Goths
whose mightiest monarch he had been, we can the more easily
understand how readily, later, the slaughter of the Harlungs
came to be attributed to the typical tyrant king, and how this
slaughter ceased to be the righteous retribution of an outraged
god, and became the act of a grasping and avaricious tyrant.
1 Enituit enim Hamalus felicitate, Ostrogotha patientia, Athala mansuetudine,
Winitarius cequitate, Unimundus forma, Thorismuth castitate, Walamer fide,
Theudimer pietate. Cassiodori Varies xi, 1 ; ed. Th. Mommsen.
2 Koediger in Weinholds Zeitschrift, i, 243; Jiriczek(2) r, 67.
3 Mommsen in a note to his Jordanes (p. 76) suggests that the name of
Athala has been inserted in the text of Cassiodorus, by error, for that of
Ermanaric. But this seems unlikely, for Athala's virtue mansuetudo would
hardly have been chosen as the conqueror's strong point. Athal is vouched for
by Jordanes' genealogy (xiv, 79) ; he is grandson of Ostrogotha.
3—2
36 Widsith
The fact, therefore, that in the introduction to Widsith
Eormanric is roundly denounced is not enough to prove that
introduction a later addition. It probably is so : but we shall
need more evidence than this.
Theodoric of Verona.
Last and longest remembered of the evil deeds of the
Ermanaric of story was his tyranny and treachery towards his
nephew Theodoric.
As a matter of history Theodoric was born on the day when
the news reached his father's house of the final victory which
liberated the Goths from that subservience to the Huns into
which they had fallen on the death of Ermanaric eighty years
before. But this liberty only meant, during the whole of
Theodoric's boyhood and youth, a state, half of dependence upon
the Eastern empire, half of hostility to it, always of penury.
At the age of twenty Theodoric became, by the death of his
father Theodemer1, leader of this little nation of soldiers out of
employ. In 488 he received from the Eastern emperor per
mission to try his fortune in Italy, then ruled by Odoacer and
a miscellaneous host of German mercenaries. So Theodoric
invaded Italy with the whole Ostrogothic nation, taking with
him remnants of the Rugian people under their king Frederic,
who had a blood feud to avenge upon Odoacer. After a four
years struggle, marked by his great victories of the Isonzo, of
Verona and of the Adda, prolonged by the treachery of his
allies, first of Tufa and then of Frederic, Theodoric at last slew
Odoacer in his palace at Ravenna on March loth, 493. Then
followed his thirty-three years of just and peaceful rule. " He
cared for justice above all things, and was a true king2."
But Theodoric had not been dead ten years when a spas
modic revival of the strength of the Eastern empire led to a
war between the Gothic nation and their Rugian allies on the
one hand, and, on the other, the " Roman " army of mercenaries
drawn from three continents, and led by two great military
1 M.H.G. Dietmar; O.N. pioftmarr. 2 Procopius, Bell. Gott>, i, 1.
Stories known to Widsith: Gothic heroes 37
geniuses, first Belisarius, then Narses. Eighteen years of in
cessant fighting followed. The Goths were hampered at the
outset by the incompetence of their leaders : and all the heroism
of later chiefs, when war had brought the right men to the front,
failed to redress the balance. Five times was Rome taken and
retaken : twice the Goths wore out their strength in unsuccessful
attacks upon the massive Aurelian walls, before the last great
pitched battle took place at the foot of Mount Vesuvius. The
Byzantine historian's account of this reminds us of the Old
English lay of Maldon. The Goths dismounted for the fight,
and the last Gothic king placed himself, with his shield-bearers,
in the forefront of the battle, changing his shield as often as it
became too heavy, from the weight of the enemy's javelins fixed
in it. When at last he was slain, in the moment of so changing
his shield, the Goths did not cease from fighting. All that day
and the next the battle continued: on the third day the
survivors asked and received permission to leave Italy, and
settle in some barbarian land, far from the power of Rome1.
Bat it took two years more to root out the isolated Gothic
settlements and garrisons scattered throughout Italy, and this
last struggle was lengthened by an intervention in favour of the
Goths made by the Alamanni. The Alamanni were bound by
ancestral ties of gratitude to Theodoric2, and now they left
their homes around the Lake of Constance to aid the Goths3
and to share in the plunder of Italy. Their intervention
failed : as so often later, both sword and plague wasted the
German invaders : but it is a very probable conjecture that the
last remnants of the Goths accompanying their allies took refuge
in the land of the Alamanni. Certainly it is there that we find
some of the earliest and most frequent traces of Gothic saga4,
and it is probably not too bold to regard the constant occurrence
of the name Amalunc around the Lake of Constance, in the
books of the Confraternities of St Gall and Reichenau, as a
trace of an actual Gothic element in the population.
1 Procopius, Bell. Gott. iv, 35; cf. Dahn, n, 176-242.
2 See Cassiodori Varies, n, 41 (letter of Theodoric to Clovis interceding
for the Alamanni).
8 See Agathias, n, 6-7.
4 See Uhland's Essay on Dietrich von Bern (esp. p. 379, note 1), in the
Schriften zur Geschichte der Dichtung, vin, 334, etc., Stuttgart, 1873.
38 Widsith
From these highlands of South Germany the story of
Theodoric spread gradually to the coast tribes of the North1,
but as it passed from tribe to tribe it became strangely altered.
For Theodoric's work was, as we have seen, undone within a
lifetime, his countrymen driven out of the land he had won for
them, and Italy, which he had ruled and protected so well, left
a prey to the ferocious Alboin and his robber dukes. Poetry
seized upon this tragic aspect of Theodoric's career. He is the
greatest of warriors, yet fortune always robs him of the fruits of
victory : so that the traditions of North Germany and the epics
of South Germany represent the conqueror of Italy as a fugitive
who only enters into his kingdom after thirty years' exile, after
having lost all save one of his vassals.
So one of the most successful figures in all history came to
be the type of endurance under consistent and undeserved
misfortune. Dietrich's lament for his fallen vassals is one of the
most pathetic things in the literature of the Middle Ages :
Owe daz vor leide niemen sterben ne mac ! 2
A northern poet brought together Guthrun and Thiothrek
as the most sorely tried of men and of women :
Never, says Guthrun when falsely accused to her husband Atli of
unfaithfulness with Thiothrek, Never did I embrace the king of hosts,
the blameless prince ; all other were our words as we two bent our heads
together, bewailing.
N£ ek halsa]>a herja stille,
J9for 6neisan eino sinne:
a]>rar v$ro okrar spekjor,
es vit hgrmog tvau hnigom at runom3.
Indeed, so unlike is the career of Dietrich von Bern to that
of the historic Theodoric that W. Grimm went so far as to deny
their identity4: two originally quite distinct figures had, he
thought, been confused. Yet the transition from the hero of
history to the hero of story is quite intelligible, if we do not
allow ourselves to forget the later fortunes of the Gothic race.
1 The spreading of tradition must have been facilitated by the intercom
munication of mercenary bands. We meet members of North-Sea coast tribes
serving in Italy a generation after Theodoric's death (Agathias, i, 21).
2 Nibelungenlied, ed. Bartsch, 2323.
3 Gu\>runarkvi]>a in, 4 (Symons, Edda, i, 410).
4 Heldensage, 344. The mythical theory of the Dietrich story was elaborated
by Halm, Sagenwissenschaftliche Studien, Jena, 1876, pp. 66, etc.
Stories "known to Widsith: Gothic heroes 39
It is true that popular poetry does not deal in the " abstractions
and generalizations of philosophical world history," and to
suppose that in the misfortunes of the Theodoric of saga the
poets consciously sought to allegorize the subsequent mis
fortunes of his people is to misunderstand the spirit of German
heroic poetry, which did not work things out in this way1. But
the Gothic stories were probably in the first place carried
beyond the confines of Italy by Gothic refugees. It would be
natural for these men to emphasize the tragic side of Theodoric's
story, and, exiles themselves, to represent their national hero as
an exile. Tradition never preserves an accurate chronology,
and Theodoric's many years of struggle, followed by thirty years
of peace, might easily become thirty years' disaster, followed by
a few years of belated victory. And so the story of Dietrich
von Bern might grow quite naturally from that of the historic
Theodoric.
On the other hand, if the events of his life are distorted, the
character of the Theodoric of history — just, sometimes stern, on
rare occasions ferocious, is reflected by later tradition with con
siderable accuracy. Contemporary Roman historians, and those
of the next generation, had every reason to bear hard upon the
memory of the Arian barbarian, at the overthrow of whose
kingdom they rejoiced. Yet they agree in representing him as
one who loved justice, and held the laws sacred : who, though
he would not take the title of Roman Emperor, and was content
to be called reiks as the Barbarians call their chiefs, was second
to none of the best in the line of Emperors — a terror to his
foes, loved and regretted by his subjects, Goths and Italians
alike 2.
The main features of Dietrich von Bern are here.
To a certain extent we can trace the development of the
historic Theodoric into Dietrich von Bern. The earliest
tradition must have followed fact in representing Odoacer as
1 Compare W. Muller, Myth. 183, " In der Sage bekam das Andenken an
die letzten Geschicke der Goten natiirlich das Ubergewicht...." Against this it
is urged by Jiriczek(2) 130, "mit allgemeinen blassen Abstractionen operiert
die Sagenbildung nicht." Jiriczek is undoubtedly right: yet national history
must unconsciously colour national tradition.
2 Procopius, Bell. Gott. i, 1.
40 Widsith
Theodoric's adversary. A record of this stage of the story is
preserved in the Hildebrand Lay. "He (Hildebrand) fled
from the enmity of Odoacer with Theodoric, and his warriors
many1." But tradition tends to make all its heroes contempo
rary, and so Theodoric became the nephew of the tyrant king
Ermanaric, whilst Odoacer first remained in the background as
the instigator of trouble2 and finally disappeared altogether.
Sifeca, Sifka or Sibech, the evil genius of Ermanaric in the
murder of the Harlungs, is made responsible for Ermanaric's
cruelty to Theodoric also.
Theodric in Widsith.
The author of Deor, to judge from the way in which he
introduces their names, knew of Eormanric as the foe of
Theodric. On the other hand, it has come to be a dogma that
Theodoric of Verona is unknown to Widsith, and that the
Theodric who is twice mentioned in the poem, first in the
catalogue of kings, and then among the champions of
Ermanaric, is that Theodoric the Frank, the son of Chlodowech,
who grew to be a hero of story second only to Dietrich von
Bern.
This Theodoric the Frank was an illegitimate son : there
were difficulties about the succession and quarrels with his
kin, followed by a victorious reign8. Not dissimilar was the
history of Theodebert his son4. From these events grew the
great Middle High German stories of Hug-Dietrich and his
son Wolf-Dietrich, stories which turn upon the reproach of
illegitimate birth attaching to Wolf-Dietrich, the jealousy of his
half-brothers, stirred up by the wicked Sabene, the allegiance
of the faithful Berchtung and his sons : Sabene and Berchtung
being contrasted much as are the false Sibech and the true
Ekkehart in the story of the Harlungs.
1 floh her Otachres nid,
hina miti Theotrlhhe, enti sinero degano filu.
See Koegel Ltg. i, 1, 216.
2 As in the Wurzburg Chronicle. Yet perhaps this reflects a confused
tradition. On the other hand it is conceivable that, in the Hildebrand Lay,
Odoacer is not the tyrant, hut the tyrant's counsellor.
3 See below, Chap, iii : Theodoric the Frank.
4 For Theodebert see Agathias, i, 4 ; Gregory of Tours, Hist. Franc. , passim.
Stories known to Widsith: Gothic heroes 41
There can be no doubt that in Widsith 1. 24 (\Jeodric
weold Froncum) the reference is to Theodoric the Frank, the son
of Chlodowech. It has been too readily assumed that the
later allusion to a Theodoric in 1. 115 refers also to the Frankish
king, and that Theodoric the Goth is left without mention in
Widsith1' from which it has been argued that Theodoric's
connection with the Ermanaric story is later, and dates from
the eighth century. Yet the assumption that the Theodric of
1. 24 is identical with the Theodric of 1. 115 is a perilous one.
The first mentioned Theodric is king of the Franks ; the latter
Theodric is a champion at the court of Eormanric. Besides, it
is dangerous to argue from one section of Widsith to the other:
the two sections may have been originally independent poems2.
The only reason for identifying the Theodric of 1. 115 with
Hug-Dietrich or Wolf-Dietrich is the fact that his name is
coupled with that of Seafola, in whom Miillenhoff saw the Old
English equivalent of the treacherous Sabene of Middle High
German story. But it was overlooked that there are two heroes
of the name Sabene : the one a faithless retainer of Hug-
Dietrich, the other — Sabene of Ravenna — the faithful retainer
of Dietrich von Bern, often referred to in Dietrichs Flucht3.
The coupling of Seafola with Theodric is then equally
appropriate, whether that Theodric be the Goth or the
Frank, and gives no evidence as to which of the two heroes the
poet had in mind. But the balance of probability is in favour
of the Ostrogoth, for several reasons :
(1) Tradition would easily make Ermanaric's kinsman
Theodoric into one of his retainers, and we know that it ulti
mately did so : but no satisfactory explanation has been given
which would account for the mention of Theodoric the Frank
among Ermanaric's household4.
(2) The omission of the name of Theodoric the Ostrogoth
from a list of Gothic champions would in any case be strange.
1 Koegel, Ltg. i, 1, 149-51; Symons in Pauls Grdr.^) in, 691; JiriczekyTi.
2 I hope to show below (Chap, iv : end), that they almost certainly were.
3 Dietrichs Flucht is a late poem : but Sabene's name probably represents
an early tradition ; for after the name had become current as that of a traitor
it would hardly have been chosen by a poet adding new characters to his story,
for a faithful retainer.
4 See note to 1. 115.
42 Widsith
(3) Although the Angles certainly knew something of the
conquests of Theodoric the Frank, we have no evidence that
the romantic story of Hug-Dietrich, Wolf-Dietrich and Sabene
ever became a part of the Anglian stock of tales. The stories
of Theodoric the Ostrogoth, however, did form part of this
stock.
For there is sufficient evidence that Theodoric the Ostrogoth
was known, and well known, to the Anglian poets. " Theodric
held for thirty winters the city of the Maerings: to many
was that known" says Deor.
The reference is obscure: we should expect rather "was
exiled from " than " held " : but the very allusiveness of the
poet seems to show that he expected his hearers to know the
story well. In the Waldere fragment1 we have another
reference to Theodric, this time in the character which he plays
so largely both in High and Low German story : that of the foe
of monsters and giants. A speaker, probably Guthhere, is
referring to a famous sword, perhaps "Mimming":
I wot that Theodric thought to send it to Widia himself, together
with much treasure... forasmuch as Widia, the son of Weland, kinsman
of Nithhad, released him from sore straits : through the domain of the
giants he hastened forth.
In the Anglo-Saxon Martyrology we are told how Theodericus
was hurled to torment down a crater of the Lipari Isles — a story
which is doubtless due to ecclesiastical hatred of Theodoric
as an Arian heretic, the slayer of Boethius, and as a hero of
secular song : the same hatred to which we owe the conclusion
of the Thidreks saga, where the Fiend, in the form of a black
horse, runs off with Thithrek. When the Anglo-Saxon compiler
adds \cet wees Theodoricus se cyning ]>one we nemnaft ipeodric*,
this seems conclusive evidence for the existence of vernacular
traditions respecting Theodoric in England at the date when
the Martyrology was written ; that is to say the ninth
century.
1 The evidence of this would of course be worthless if, as Koegel (Ltg. i, 1,
235-41) and Binz (P.B.B. xx, 217) suppose, the Waldere is merely a literal
transliteration from the O.H.G. But Binz himself has shown how insufficient
is the evidence brought forward by Koegel to support this theory.
2 Old English Martyrology, ed. Herzfeld, E.E.T.S., p. 84.
Stories known to Widsith: Gothic heroes 43
It has further been pointed out1 that Ida of Bernicia named
two of his sons Theodric and Theodhere : Theodhere (Diether)
was, in story, the name of Theodric's young brother, whose
premature death at the hands of Theodric's former friend and
follower, Widia (Wittich), was the most cruel of the blows which
fell upon Dietrich von Bern. Perhaps this combination of
names in the Northumbrian genealogy is a mere coincidence :
if Ida had the story in mind in naming his sons it must have
spread north very early ; for Ida was ruling in Bernicia barely
a generation after the death of Theodoric the Great, and some
years before the last of the Ostrogoths were expelled from
Italy.
But other personal names seem to confirm a knowledge of
the Theodric story in England2, and when we find the informa
tion volunteered in the Alfredian translation of Boethius, se
peodric wees Amulinga, we can hardly hesitate to attribute this
to epic tradition. Add the fact that Theodric's henchmen, the
Wylfingas — the Wiilfinge of High German tradition — are also
well known in English story3, and it seems ultra-scepticism
to doubt, as does Binz4, that the mighty figure of Theodoric
played a considerable part in the Anglian lays. Indeed the
evidence for the popularity of Theodoric the Goth is far better
than the evidence for the popularity of Beowulf the Geat.
For a passing allusion, which the poet evidently expects his
readers to understand, is better evidence of popularity than a
story told at full length : and we have not one allusion only,
but several.
When then, in Widsith, we find the name Theodric in
a place where Theodoric the Goth is wanted, and where
1 Brandl in Pauls Grdr.($ n, 1, 953. It is so natural for kinsmen to bear
names containing the same element that this may well be a mere coincidence.
2 Lib. Vitae, ed. Sweet in Oldest Eng. Texts, p. 159, line 212. A collection
of the early allusions to Theodoric in England will be found in Miillenhoff s
Zeugnisse u. Excurse, v, 1, in the Z.f.d.A. xn, 261-2. Miillenhoff notes the
occurrence of the name Omolincg, Omulung in two very early charters.
The signatures are respectively of the years 692 and 706, or perhaps some
years later, for the name occurs in the endorsement (see Birch i, 76 and 116).
The signatory Omolincg, Omulung is clearly one and the same, the abbot
of some monastery presumably in the district of Worcester. It is conceivable
that Omulung is a foreign cleric, an immigrant from the great abbeys of
St Gall or Beichenau, where we know the name to have been common.
3 See note to 1. 29. * P.B.B. xx, 212-214.
44 Widsith
Theodoric the Frank would be a most unwelcome intruder,
there is surely no reason why we should abstain from the
obvious interpretation of Seafola ond Ipeodric as Theodoric the
Goth and his retainer Sabene of Ravenna1.
Whether Anglian story at this date represented Ermanaric
as the relentless foe of Theodoric is another question. Probably
not : for since in Widsith Ermanaric is not yet thought of as
ruler of Italy, but governs the Goths in their old home by the
Vistula, there is no reason why Theodoric's invasion of Italy
should be regarded as directed against him. In Deor the
mention of the tyranny of Eormanric immediately after the
reference to Theodric probably points to a connection in story
between the two : but of this we cannot be certain.
Attila.
j?Etla weold Hunum.
No very considerable lapse of time would be needed before
tradition brought together Ermanaric and Theodoric as con
temporaries. For the intervening years had been a blank in the
history of the East-Goths ; years spent in the service of their
Hunnish overlords ; and the great deeds which had been
achieved during this time had gone to the credit, not of the
Gothic name, but of Attila, king of the Huns.
Two distinct traditions concerning Attila can be traced
among the German races. On the one hand South German
tradition represents him as a great and hospitable king, the
holder of a court frequented by valiant knights. Like Arthur
in a similar position, he is overshadowed by his knights and by
his queen, till he becomes a commonplace, and sometimes a
1 See Heinzel, Osgotische Heldensage, 8. Heinzel has noticed the double
occurrence of Sabene: he and Holthausen (Beowulf, p. 125) are, I believe, the
only critics who interpret the peodric of 1. 115 as Theodoric of Verona. At an
earlier date (Hervararsaga in W.S.B. cxiv, 50) Heinzel had given his adhesion
to the common belief that the reference was to Hug- or Wolf-Dietrich. Of
course, if the hypothesis of Bugge (H.D. 71, 167-8; P.B.B. xxxv, 267) and
Miiller (202) is correct, that Wolf-Dietrich and Sabene are originally figures of
Ostrogothic tradition, this would account for the way in which they are
introduced into Widsith. But this theory has not been proved.
Stories known to Widsith: Gothic heroes 45
weak figure. But he is always kindly and hospitable1. On the
other hand there is the Atli of Scandinavian story, grim,
covetous, terrible, who murders the Niblungs, the brethren of
his wife, and perishes at length through the vengeance of that
wife.
When we remember that in the great battle on the Mauriac
plain almost the whole inland Teutonic world was ranged, the
one half as Attila's foes and the other half as his henchmen, we
can understand how this double tradition grew up in German
lands. To the Rugian, the Gepid, the Thuringian and the
Eastern Frank, but above all to the Ostrogoth, Attila would be
the great king in whose hall their chiefs feasted, and following
whose banners they had gained all that a German warrior most
loved —
sweord and swatigne helm, swylce eac side byrnan
gerenode readum golde.
To them Attila would be indeed the " little father."
To the Visigoth on the other hand, the Salian Frank, and
the Burgundian, Attila would be the grasping "tyrant of the
world, making war without a quarrel and thinking any crime
allowed him2." The Atli of the North seems to be derived,
through various stages of Low German saga, from the point of
view of the Salian Franks and other nations who fought against
Attila in the great battle3, whilst in the South German Etzel
we can trace quite clearly the benevolent overlord of Ostrogothic
tradition. Not that the old hostility of Goth and Hun was
forgotten in this South German tradition4: on the contrary we
shall see that in the person of the old national hero Widigoja
(the Wittich of High German story, the Wudga or Widia of
Anglo-Saxon tradition) recollections of the old struggle with
the Huns in the days of Ermanaric were mingled with a vague
remembrance of the death of Attila's sons in the war of libera
tion eighty years after Ermanaric's death. But in Southern
story the Gothic cause grew to be identified with the tyrant
1 See Vogt, Z.f.d.Ph. xxv, 414; Koegel, Ltg. i, 2, 283; Hodgkin's Italy, n,
176; Thierry, Histoire d' Attila, 1864, n, 224, etc.
2 Jordanes, ed. Mommsen, xxxvi (187). The words are put into the mouth
of the Koman envoys at the Visigothic court.
3 Pauls Grdr.ty m, 700.
4 See Appendix (I), Ermanaric as the foe of the Huns.
46 Widsith
Ermanaric, whilst Attila, Etzel, was glorified as the patron and
helper of Dietrich von Bern — no very strange part for him to
play, when we remember that Attila actually was the leader
and friend of Theodemer, Theodoric's father. So that in that
very South German tradition which was most closely in touch
with Ostrogothic story, the sympathy of the poets was rather
with the Huns than with the Goths. In Low German
tradition, on the other hand, Attila was the tyrant foe : and
the form which his name takes in Widsith — ^Etla, not Etla —
points conclusively to Low German tradition1. In Widsith,
therefore, the struggle between Goth and Hun is looked upon
from the Gothic side : the scene of the conflict is the old home
of the Goths by the Vistula, where ^Etla, as the typical Hunnish
chief, is, despite chronology, opposed to Eormanric and his
men. Chronology has been also disregarded in the list of
Ermanaric's champions : earlier heroes like Eastgota or
Wudga have been drawn into the great conflict with the
Huns.
The tradition of this conflict has also been preserved to us
in the Icelandic : in the prose of the Hervarar saga and in the
verse lay of Hloth and Angantyr : a fine poem, but so allusive
and so corrupt that without the help of the prose saga we should
hardly be able to understand it. The Hervarar saga is the life
story of a sword " Tyrfing," the work of dwarfs, which is fated,
every time it is drawn, to be the death of a man. The tale of
the race of heroes through whose hands the sword passes is
traced from generation to generation till we come to Angantyr,
third of the name, who has inherited the lordship of the Goths
and the sword Tyrfing from his father, the grim king Heithrek.
As Angantyr is drinking the funeral feast over his father,
Hloth, a bastard son of Heithrek by the daughter of Humli,
king of the Huns, comes riding from the East to claim one half
of the inheritance. He speaks in the poem :
" Half will I have of all that Heithrek owned — cow, calf, humming
hand mill... half the war gear, the land, and the shining rings."
" The white shield must be cloven," Angantyr replies, " and cruel spear
clash with spear and many a man fall on the grass, ere I will halve mine
inheritance with the son of Humli, or divide Tyrfing in twain."
1 See note to 1. 18.
Stories Tmown to Widsith: Gothic heroes 47
Yet Angantyr offers mighty treasures to Hloth, men, horses,
maids, necklaces, silver and gold, "and the third part of the
Gothic folk shalt thou rule alone." At this old Gizur, the
foster father of Angantyr, is wroth and grumbles " That is an
offer meet to be accepted by the son of a bondwoman, the son of
a bondwoman though born to a king " : and with this taunt all
thought of peace is gone. Humli calls out the Hunnish host,
from the boy of twelve winters and the foal of two winters, in
order to vindicate his grandson's right, and they fall upon the
Gothic borders, and first upon the amazon sister of Angantyr,
Hervor. Hervor is slain, and Ormar the master of her host
comes with the news to Angautyr, "drenched is all Gothland
with the blood of men." Angantyr sends Gizur to challenge
the Huns to battle at Dunheath ; the challenge is delivered and
Gizur is allowed to ride away in safety, to tell to Angantyr the
vast numbers of the Hunnish hordes1. From the prose saga we
get a lengthy account of the battle, waged for ten days2. But
the poem omits all account of this, and passes at once to
the last scene of all, as Angantyr stands over his dying half-
brother : " 111 is the doom of the Fates."
The connection between this story and the Ermanaric
catalogue in Widsith was noticed by C. C. Rafn in his edition
of the Hervarar saga3 half a century or more ago. A few years
later another scholar4 pointed out that the Hltye and Incgerityeow
of our poem were Hloth and Angantyr.
It seems clear that the Hervarar saga and the Ermanaric
catalogue in Widsith go back to a common tradition, which had its
roots in the early history of the Huns and the Goths. Whether we
can fix any precise event in that early history as the origin of
the story is another matter. Heinzel has tried to do this, and
has drawn attention to a number of remarkable parallels between
the incidents preceding the battle on the Mauriac plain, and
1 This part of the story Saxo knew, and incorporated into his History of the
Danes (ed. Holder, 154-5).
8 Fornaldar Sqgur, 1829, i, 503-8. The Lied von der Hunnemchlacht will
be found in the Eddica Minora of Heusler and Banisch, Dortmund, 1903 ; also
in the text of the Hervarar saga.
3 In the Antiquity's russes Eafn identified Ormar and Wyrmhere : his other
attempts at identification (e.g. Angantyr and Eastgota) were less fortunate. See
notes to Wyrmhere (1. 116) and Hlfye (1. 119) below.
« See-note to 1. 116.
48 Widsiih
the incidents in the closing chapter of the Hervarar saga1.
But in all probability no one specific battle gave rise to the
tradition. Into the ten days' battle of Dunheath Norse poetry
has probably compressed the century long struggle of Goth and
Hun : and Widsith is probably right in pointing to a succession
of fights rather than any one pitched battle. For popular
tradition will easily turn a desultory conflict into a single
dramatic encounter, but hardly the reverse2.
And the fact that in the Norse story Atli is not the leader
of the Huns, is strong evidence that the tradition does not
derive from Attila's great reverse on the Mauriac plain. For
to Teutonic tradition Attila is essentially the Hun. That he
should be introduced into Widsith is therefore natural, even
if the fighting on the Vistula reflects those ancient battles
which preceded the fall of Ermanaric, long before Attila's days.
But when in the Hervarar saga we find that the Hunnish leader
is not Atli but Humli, we may be sure that this is an original
trait of the story. Attila could hardly have lost his place as
Hunnish leader in the battle of Dunheath, had he ever rightly
possessed it8.
Wudga and Hama.
Ne woeran ^cet gestya ]>a scemestan,
}>eak\>e ic hy anihst nemnan sceolde...
Wrceccan ]>cer weoldan wundnan golde,
werum and wifum, Wudga and Hama.
To two heroes, met during his wanderings, Wudga and
Hama, our poet does special honour. The word wrcecca, which
is applied to them, comes later, in High German, to signify
simply champion, or warrior. But in Old English it has not
lost its significance of " exile "; and in this representation of
Wudga and Hama as strangers within the Gothic borders we
have perhaps an indication that, at the time when Widsith was
1 Hervararsaga, esp. pp. 50, etc. Heinzel's theory is discussed and
rejected by Veselovsky in the Bussian Journal of the Ministry of Public
Instruction, May, 1888, pp. 74-90, "Short notes on the Byliny." See
summary by V. Jagic" in the Archiv fur Slavische Philologie, Berlin, 1888,
xi, 307.
2 See Jagtf, p. 308.
8 For reasons to the contrary see, however, Heinzel, Hervararsaga in W.S.B.
cxiv, 79.
Stories known to Widsith: Gothic heroes 49
written, it was still felt that they did not rightly belong to the
Ermanaric cycle : much as Sir Tristram, though he visits the
court of Arthur and is enrolled a knight of the Table Round, is
never quite at his ease there, but betrays that he has stepped
out of a story originally distinct. It is probable however that
wrceccan means more than this, and that in the stories known
to Widsith, Wudga and Hama were outlaws.
Wudga is probably a more ancient figure of Gothic tradition
than even Ermanaric.
Vidigoia1 is mentioned by Jordanes as one of the Gothic
heroes whose memories were preserved in popular song3: in a
later chapter a locality is mentioned " where of old Vidigoia,
the bravest of the Goths, fell by the wiles of the Sarmatians3."
In the course of generations the Huns, as the historic foes of
the Goths, naturally took the place of the Sarmatians, whilst
Widigauja came to be drawn into the cycle of two distinct
Gothic kings, whose fame eclipsed his own. As the foe of the
Huns he was naturally connected with Ermanaric4, whilst tales
of adventures with giants and monsters made him the associate
of Theodoric. These tales are preserved both in the Middle
High German and in the Thidreks Saga: we have seen that
the Old English Waldere alludes to the reward Widia received
for releasing Theodric from straits, presumably among the
giants and monsters. Thus Widigauja, Wittich, owes fealty
to two kings ; and when later tradition made these kings deadly
foes, his position became a difficult one. True, story represented
him as having entered the service of Ermanaric, with the con
sent of his former lord, Theodoric, before their quarrel5. But,
1 Corresponding to a presumed Gothic * Widigauja.
2 Ante quos etiam cantu maiorum facta modulationibus citharisque canebant,
Eterpamara, Hanale, Fridigerni, Vidigoia et aliorum, quorum in hac gente
magna opinio est, quales vix heroas fuisse miranda iactat antiquitas. Cap. v
(43) ed. Mommsen.
3 ingentia si quidem flumina, id est Tisia, Tibisiaque et Dricca, transientes
venimus in loco illo ubi dudum Vidigoia Gothorum fortissimus Sarmatum dolo
occubuit ; indeque non longe ad vicum, in quo rex Attlla morabatur, accessimus.
xxxiv, (178). The spot is probably in the centre of what is now Hungary.
The rivers mentioned are not easy to identify, but one of them is apparently
the Theiss. See Wietersheim, Geschichte der Volkerwanderung , 2 Aufl. herausg.
Dahn, n, 230-31.
4 See Miillenhoff, Z.f.d.A. xn, 256.
6 Alpharts Tod, 26; cf. Jiriczek(2) 307.
c. 4
50 Widsith
even so, his honour was not, in the High German versions,
saved, and Vidigoia Gothorum fortissimus becomes " Wittich
the untrue1."
One of the worst checks received by the historic Theodoric
in the long struggle which made him master of Italy was due
to his having first received into his service, and having then
been deserted by, a general of his adversary's, Tufa. Perhaps
this double treason came to be ascribed to Wittich2. Even his
valiant deeds against the old enemies, the Huns, stood Wittich
in little stead. The legend which made him the slayer of the
sons of Attila is probably, as Heinzel has pointed out, an echo
of the two great battles in which the Goths threw off the yoke
of the Huns, defeated the sons of Attila, and slew one of them ;
the battle of the river Nedao in the year 454 3, and the subse
quent and final victory, probably of the same year. The historic
victors in these battles were the father and uncles of Theodoric,
but, as the national champion against the Huns, it was im
possible to keep Widigauja out of the business. When, however,
later story made Attila Theodoric's refuge from the treachery
of his uncle Ermanaric, the slaying of the children of Attila
became one of the worst of the many evil acts of Wittich.
But only one out of many ; for Wittich came to be a sinister
figure : the merciless slayer of youthful princes. The death of
one of these, Nuodunc, is referred to several times in the
Nibelungen Lied: it is one of the old sorrows lying in the
background, when the Burgundians are hospitably entertained
1 da ndmen si den ende
von des ungetriuwen Witegen hende.
Rabenschlacht, stanza 364.
2 Dietrichs Flucht, 7133 etc., 7712 etc., cf. Max Eieger in Z.f.dM. i, 233.
The connection between the treason of Wittich and that of the historic Tufa is
often spoken of as certain (e.g. by Jiriczekfj) 133 ; (2) 136). But there are serious
difficulties : cf. Boer, Ermanarich, 88 ; Miiller, 160. Apparently it is only in
the latest versions of his story that Wittich becomes a traitor. We should have
to assume, either that a tradition of Tufa's treachery survived independently
to a late date, and was transferred to Wittich after his degradation ; or that,
under learned influence, the historic fact of Tufa's treachery was attributed to
the traditional Wittich, by some such combination of history and legend as
has been suggested for the old French epic by Be'dier (Les legendes epiques,
108 etc., 399 etc.).
The connection sometimes suggested with the historic Witigis (e.g. by
Abeling, Das Nibelungenlied, 1907, p. 221; Matthaei in Z.f.d.A, ZLVI, 51 etc.)
eeems very doubtful.
3 Hodgkin's Italy, n, 192; ra, 14.
Stories known to Widsith: Gothic heroes 51
by the Margrave Riidiger1 and his lady. But the story is not
told at length in any extant High German poem : for in High
German story the place of Nuodunc was taken by the young
Alphart, whom Wittich does to death with the help of his
friend Heime, and that too by a foul stroke2. As well as
Nuodunc, Alphart and the sons of Attila, Wittich slays the
young brother of Dietrich, Diether der vurste wolgetdn3. But
vengeance is at hand in the person of Dietrich himself. He
gives chase to Wittich, whom neither taunts, nor the example
of his uncle Reinolt, can persuade to stand his ground. But
Dietrich gains on him, and the sea is in front. Wittich has
already rushed into the sea when "at that moment came a
mermaid.... She took the mighty warrior, and carried him with
her, horse and all... with her, down to the bottom of the sea4."
It is unusual for the god in the machine to appear in order
to save the traitor from the vengeance of the hero : and may
we not safely see in Wittich's end a survival from the story of
the valiant champion of earlier date5?
We have no reason to think that degradation overtook
Wudga in any of the English stories: for even the conti
nental Saxon versions, as translated in the Thidreks saga, kept
his honour unimpaired. In the Thidreks saga Vithga passes,
with Thithrek's consent, into the service of Erminrek, whilst
these heroes are yet friends. He warns his old master of the
evil which is compassed against him by Sifka's wiles : but when
Thithrek tries to regain his own by force Vithga has, against
his will8, to bring up his liegemen to fight under Erminrek's
banners. Deserted by Sifka, he rallies the host, slays first
Nauthung, then the sons of Attila, and at last, after warning
him in vain, Thithrek's young brother; and falls a victim to
Thithrek's revenge. The interest to the Saxon poet lay in the
1 Der Nibelunge Not, ed. Bartsch, 1699: cf. also 1903, 1906, 1927. For
other references see Grimm, H.S. 101 (third edit. 112).
2 Alpharts Tod, 289. The version is confused, fragmentary and interpolated.
3 Rabenschlacht, stanza 453, etc.
4 Eabenschlacht, stanza 964-5 ; cf. 966-974.
5 As is noted by Grimm, H.S. 360 (third edit. 408) there are noble features
even about the High German Wittich, such as his lament after he has smitten
Diether.
6 Cap. 323.
4—2
52 Widsith
story of the good knight compelled to fight those who are dear
to him ; the pathos is that of the Margrave Eiidiger, in the
Nibelungen Lied, or of the closing chapters of Mallory, where
Lancelot and his men wage war against King Arthur. This
feeling receives fine expression in the preceding scene, where
the old friends, Master Hildebrand and Reinald, meet in the
moonlight by the river between the two hosts: "Then said
Hildebrand, ' Where is Vithga, our beloved friend, with his
host ? ' Then Reinald answered, ' You can see the green tent,
with the silver knob on the top of the pole. There Vithga
sleeps, and many Amlungs. And they have vowed that to
morrow they will cleave many a helm1.'"
" In Witege and Heime later heroic story has personified,
in duplicate, the faithless, venal, cold-hearted and sinister
champion2." We have seen that, with regard to Witege, this
must be understood with limitations : and equally is this the
case with regard to Heime.
There is no historical incident with which we can associate
Hama's name. The earliest references are in Widsith, where
Hama is Wudga's comrade ; the already mentioned St Gall
document of 786, in which a Heimo and his daughter Suanailta
occur ; and the puzzling passage in Beowulf:
sy)>tSan Hama aetwaeg
to )>sere byrhtan byrig Brosinga mene,
sigle and sincfset, searoniSas fealh [Grundtvig, Leo, fleah]
Eormenrices, geceas ecne rsed.
Since Hama carried off to the bright town the necklace of the Brisings,
jewel and precious work : penetrated [better, fled] the wiles of Ermanaric,
chose the eternal welfare3.
1 Cap. 326. Vithga overshadows his master, and is the real hero of much
of the Thidreks saga, as Launcelot of the Morte Arthur. Vithga takes a
robbers' castle which has defied Thithrek (cap. 84), and it is only by borrowing
Vithga's sword Mimung that Thithrek is able to accomplish his greatest
triumph, and vanquish Sigurd (cap. 222).
2 Symons in Paul's Grdr.y) in, 694.
3 Bugge (P.B.B. xii, 70) takes geceas ecne reed to be a reference to Heime's
retirement to the monastery. But this, though a motive common enough in the
later romances, seems out of place in Beowulf. It is much more probable that
it means "he died" (Mull, in Z.f.d.A. xn, 304). It is possible that the
retirement into the monastery which we find in the Thidreks saga is a weaken
ing of some more tragic end in the earlier story, just as the vengeance which
Dietrich works upon Elsan (Rabenschlacht, stanza 1120) is ultimately com
muted into retirement to a monastery. Cf. Jiriczek(2) 318. Or does the phrase
simply mean c ' he did right " (in harrying Ermanaric) ?
Stories knoivn to Widsith: Gothic heroes 53
From these last documents Mlillenhoff hazarded the guess
that Hama was the name of the treacherous retainer whose
desertion of Ermanaric is recorded by Jordanes, and that the
fraudulentus discessus was the robbing Ermanaric of his
treasure1. This seems improbable. The evidence of the
charter, far from supporting it, is against it. Is it likely that
a man named Heimo, if this had been the name of the legendary
chief who did the deed, and left his wife to bear the punish
ment, would have given to his daughter the name of that
hapless wife ?
It has generally been assumed that the reference to Hama
in Widsith points to him as still a retainer of Eormanric. But
the passage needs careful examination. Widsith travels through
the land of the Goths and visits the company of Eormanric :
then follows a long list of names, concluding with a reference
to combats with the people of Attila. Then comes a second
list of names, and it is in this second list that the names of
Wudga and Hama occur. Of them it is said that as exiles
(wrceccan) they ruled by means of twisted gold over men and
women.
Now a comparison of the more than twenty places in which
the word wrcecca occurs in Old English poetry shows it to have
a distinct and emphatic meaning : " exile," " outlaw," even
" outcast." Of course Wudga and Hama may be outlaws who
have taken refuge at the court of Eormanric, as suggested
above : but this is not the most obvious interpretation of the
lines, nor does it explain why the group to which Wudga and
Hama belong should form a separate list distinct from the
other followers of Eormanric. It seems more natural to read
the passage thus : — the poet traverses the Gothic land, visiting
first of all the court of Eormanric; then, after leaving that
court, he visits (still in or near the Gothic land) other heroes,
imagined as dwelling in exile on the outskirts of the kingdom
of Eormanric. This interpretation derives some support from
the phrase used of Wudga and Hama : " they ruled there by
means of twisted gold over men and women." To rule by
means of gold, to give gold, is in Old Germanic poetry the
1 Z.E. xin in Z.f.d.A. xn, 302-6.
54 Widsith
peculiar prerogative of the independent or semi-independent
chief. It is a phrase hardly appropriate to a retainer.
Turning to the Thidreks saga we find Heimir essentially
an outlaw champion. Whilst still a lad, he takes service with
Thithrek ; but he is exiled for a fault. He lives as a robber
and, once at least, has to save himself by flight. " Then did
Heimir say that which many men have since found to be true,
that no iron is so true to its lord as is the spur1." At length
Thithrek pardons him, and he returns to his allegiance. But
he plays only a subordinate part till the time comes when
Thithrek himself is exiled, when Heimir denounces Erminrek
to his face, escapes, thanks to Vithga's timely aid, from pursuit,
and lives thenceforth as an outlaw, harrying Erminrek's land2.
It is to be noted that he is not represented as accompanying
Thithrek in his exile, and he takes no part in Thithrek's
subsequent attempts to win back his own by force. He lives
his outlaw life for twenty or thirty years, burning and slaying
by day and by night, till at length he repents of the evil,
and enters a monastery. Here Thithrek finds him, after all
his other knights have been lost, and a fine scene ensues
between the two grey-headed men, as Thithrek slowly breaks
down Heimir's resolve, and draws him forth again into the
world3.
Our sources are, then, consistent in the view they take of
Hama. In Beowulf he is a fugitive, possessed of treasure,
robbed, apparently, from Ermanaric : in Widsith he is an exile,
possessed of treasure, and standing to Ermanaric in some
relation no further defined: in the Thidreks saga he is an
outlaw, and gains treasure by plundering Erminrek's land.
From these references one critic4 has argued that Wudga
and Hama were originally figures connected, not with Er
manaric, but with Theodoric, friends who follow him into exile
when he flees before Ermanaric. But in no version of the story
are they so represented : not in the High German, nor in the
Thidreks saga, where Vithga and Heimir, though friendly to
Thithrek, expressly do not follow him into exile. That Heimir
1 Cap. 116. 2 Cap. 288, 429. » Cap. 434.
* Boer, Ermanarich, 185.
Stories known to Widsith: Gothic heroes 55
especially should not do so is strange ; for Heimir is in the
Thidreks saga the Bedivere of Thithrek's company : " first made
and latest left of all the knights." That he should not be with
his master in his greatest need, but should choose to live as a
solitary outlaw harrying Erminrek, is not what we should expect
any story-teller to invent. Such an extraordinary trait can only
be accounted for in one way : it must be part of an original
tradition which represented Hama as thus harrying Ermanaric
on his own account, and which probably goes back to a period
before any tradition of hostility between Ermanaric and
Theodoric had grown up.
And this is confirmed by the references to Hama in Old
English poetry. In Beowulf there is no mention of Theodoric :
this indeed cannot count for much, for the allusion to Kama's
story is hasty, and no details are given. But in Widsith also
it is quite clear that Kama's exile is not connected with that
of Theodoric. For the visit of Widsith to the land of the
Goths, in which he meets with the wrceccan, Wudga and Hama,
is imagined as taking place before the evil deeds of Eormanric
have occurred. Amongst the household of Eormanric Widsith
meets the Harlung brethren, and apparently Theodoric himself.
The exile of Hama in Widsith is, then, independent of any
quarrel between Eormanric and Theodoric. More probably,
Hama is the typical outlaw of early Germanic tradition; the
forerunner of those bandit heroes whom we meet later in
Icelandic saga and English ballad. This would account for
certain rough features which characterize the Heimir of the
Thidreks saga, a slayer of merchants1 and burner of monasteries2,
of whom it is said that " had he had never so much gold and
silver, the monks would not have received him had they known
he was Heimir Studasson8." Heimir is contrasted with the
more chivalrous Vithga4: in the High German, on the other
hand, Heime is more noble than Wittich. In Alpharts Tod
a clear distinction is drawn between him and Wittich. Heime
is placed in a difficult position, but that man must be indeed
a censorious judge, who, after reading the poem through, feels
1 Cap._ 110. 2 Cap. 435. 3 Cap. 429. 4 Cap. 87, 108, 207.
56 Widsith
that his honour is besmirched1. It is Wittich who, after Heime
has stepped in to rescue him, takes an unfair advantage.
One of the best of all the old stories told in the Thidreks
saga is that of how Heimir betook himself, in his old age, to
the monastery, to atone for his sins ; surrendering to the abbot
all his wealth, his weapons, his horse and store of gold and
silver, but concealing his name. The monastery is harried by
a giant, who challenges the monks to fight out the quarrel.
The monks can find no champion, though they promise a
mansion in paradise to their representative in case of defeat.
Heimir learns the trouble from his fellow monks in the chapter
house, and volunteers to meet the giant ; rather to the distress
of the abbot, who does not approve of a monk of his being a
fighting man. Heimir asks for his weapons back : the abbot
answers falsely :
" Thou shalt not have thy sword : it was broken asunder and a door
hinge made of it here in the monastery. And the rest of thine armour
was sold in the market-place." Then spake Heimir "Ye monks know
much of books, but little of chivalry : had ye known how good these
weapons were, ye had never parted with them." And he sprang towards
the abbot, and took his cowl in both hands, and said " Verily thou wast
a fool, if no iron would suit thee to furnish thy church doors, but my
good sword Naglhring, which has cut asunder many a helm, like cloth,
and made many a son of the giants headless : and thou shalt pay for it."
And he shook the cowl, with the head inside, so hard that four of the
abbot's teeth fell out ; three on to the floor, and the fourth down his
throat. And when the monks heard mention of Naglhring, then they
knew that it was Heimir Studasson, of whom they had oft heard tell.
And they were sore afraid, and took the keys, and went to the great chest
where all his weapons were stored. One took his sword Naglhring, the
second his hauberk, the third his helm, the fourth his shield, and the
fifth his spear. And all these weapons had been so well stored that
they were no whit worse than when he parted with them.
And Heimir took Naglhring, and saw how fairly its edges and its
gold ornaments shone : and it came into his mind what trust he had had
hrits edges each time that he should fight. And as he thought of many
a happy day, and how he had ridden out with his fellows, he was first
red as blood, and then pale as a corpse. And he kept silence for a time.
After that he asked where was his horse Rispa. And the abbot made
answer " Thy horse used to draw stones to the church : he has been dead
many a year2."
In this picture of Hama become a monk, standing before
the abbot and the assembled chapter, dreaming of the old days
when he rode out with Ermanaric and Theodoric, there is
1 Cf. especially stanzas 280, 287. 2 Thidreks saga, 431.
Stories known to Widsith: Gothic heroes 57
something typical of the decline and end of the Gothic stories.
The old heroic world disappeared before the mediaeval church
and mediaeval chivalry1.
In England there is evidence that even before the Norman
conquest the heathen heroes were being forgotten ; whilst we
can trace in the Latin chroniclers2 indications that a new heroic
world, the heroes of which were the Christian kings of England,
was gradually displacing the old. In France Charlemagne, the
champion of Christendom, supplanted those older heroes whose
songs had been chanted in the halls of Karl the Great. In
Scandinavian lands, it is true, the old stories survived: but
passing through the Viking age they developed into something
different in spirit from the songs of the earlier migrations.
Still more was this the case in continental Germany. The
old names survived, and some fragments of the old stories,
but the spirit that breathed through them was different.
1 For the last English reference to Wudga and Hama see Appendix K.
2 Especially William of Malmesbury, and, to a less degree, Henry of Hunt
ingdon.
58 Widsith
Burgundian heroes: The Burgundians at Worms.
The Burgundians originally dwelt in Eastern Germany1.
Broken, about the middle of the third century, in a quarrel
with the Gepidae, they made their way across Germany to
the banks of the Rhine, and before the end of the century
they were threatening the Roman frontier from across the
river: if we can believe Orosius, — which we cannot — they
had, by the year 374, so far recovered from their early dis
asters as to number 80,000 fighting men2. Shortly after Alaric
and his Goths had sacked Rome, Gundahari, King of the Bur
gundians, claimed his share in the plunder of the empire by
setting up a sham emperor, Jovinus. " Every circumstance,"
says Gibbon, " is dark and extraordinary in the short history of
the reign of Jovinus8." The all-powerful Goths seem to have
supported him. But he proved unsatisfactory, and we find
Attawulf, the successor of Alaric in the leadership of the
Goths, undertaking "if fair and honourable terms were
offered" to send in the head of the pretender. We must
hope that Gundahari — "the glorious Gunnar" — was no party
to this buying and selling of the emperor he had made : but it
looks suspicious that the same year saw the destruction of
Jovinus, and the apparently peaceable acquisition by the Bur
gundian chief of the long-coveted territory on the left bank of
the Rhine4. So Gundahari ruled at Worms5, ze Wormez df
den sant, and this obscure incident of the puppet emperor6
1 See note to 1. 19.
2 Orosius vii, 32 (A.U.C. Mcxvm, Burgundionum quoque, novorum hostium,
novum nomen, qui plusquam octoginta millia (ut ferunt) armatorum, ripce
Rheni fluminis insederunt (see Migne (1846) xxxi, 1144). This is a misrepre
sentation, both as to the numbers of the Burgundians and as to the novelty of
their appearance, of the entry in Jerome's continuation of the Chronicle of
Eusebius, A.D. 377 (374) Burgundiorum SQfenne millia, quot numquam antea, ad
Ehenum descenderunt. (Jerome in Migne (1846) xxvn, 698.)
3 Gibbon, ed. Bury, in, 343-4. Of. Hodgkin's Italy, i, 829-30.
4 Burgundiones partem Gallitz propinquantem Rheno obtinuerunt. Jovinus
et Sebastianus fratres in Galliis regno arrepto interempti. Prosper Aquitanus,
Chron. Integrum in Migne (1846) LI, p. 591.
6 There is still a Guntersbluhm between Worms and Mainz (Jahn, 329-30).
8 It is through a fragment of Olympiodorus that we know of Gundahari's
share in the business : 'lofiivos ev MowSia/cy [read T&oyowria.K<fi] rrjs erf pas
Teppavlas, Kara <nrov8r)v Fwdp TOV 'AXavov Kal Tuvnapiov 6s (ptiXapxos ^xpTj/tdTife
ruv BovpyovvTidvwv rvpawos avTjyopetidi). Miiller, Fragmenta Historicorum
Grtecorum (1851) vol. iv, p. 61. Probably this happened in the year 411.
Stories known to Widsith: Burgundian heroes 59
would have been all history had to tell of him, had he not, more
than a score of years later, fought two pitched battles, in the
second of which he and all his warriors were slain.
Chief and Retainer in German Story.
Gundahari's death was an incident of the type which most
appealed to the poets and warriors of the ancient German
world1. How some great lord, attacked by hopeless odds, is
defended by a small group of faithful followers, who keep up a
desperate fight after their lord is slain, doggedly determined
not to leave the field on which he lies dead : this is a tale
which, in historic prose or heroic verse, in bombastic Latin or
Greek periods, or in simple vernacular dialects, is told over and
over again throughout the " Dark Ages8."
Long before, Tacitus had mentioned this loyalty of the
retainer to his lord as one of the essential features of German
character. "It is a lifelong disgrace and reproach for the
retainer to have left the field when his chief has fallen. To
defend and protect his lord, and to assign to him the honour of
his own brave deeds, is the retainer's essential pledge of
loyalty3." But the loyalty of the retainer does not as a rule
appeal very strongly to the Roman or Byzantine writer. When
the bodyguard of Odoacer all fell around their lord in the
palace hall of Ravenna, the chronicler is satisfied with the
dry entry occisus est Odoacar rex a rege Theodorico in palatio
cum commilitibus suis*, words which may record a struggle as
heroic as those in the halls of Atli and Rolf and Finn. It is
1 Cf. the section Hirdmandsdtfd og efternuele in Axel Olrik's Danmarks
Heltedigtning, 1903.
2 Compare especially the episode of Cynewulf and Cyneheard in the Anglo-
Saxon Chronicle: the Battle of Maldon : and several passages in Saxo Gram-
maticus, especially those in Book n founded on the lost Bjarkamal :
Ad caput extincti mortar ducis obrutus, ac [better at] tu
Ejusdem pedibus moriendo allabere pronus,
Vt videat quisquis congesta cadavera lustrat
Qualiter acceptum domino pensavimus [pensarimus] aurum.
(Saxo, ed. Holder, 1836, p. 66.)
Almost exactly the same phrases occur in Maldon; e.g. 11. 290-294 and 317-319.
3 Germania, xiv.
4 Fasti Vind. priores, in Mommsen's Chronica Minora, i, 320 (M.G.H.).
Interfecti sunt quivis ubi potuit reperiri says the Anon. Valesii.
60 Widsith
without much enthusiasm, though fortunately more at length,
that Agathias records the heroic death of Phulcaris — Folchere —
and his thegns. Folchere, a leader of Herulan allies, had allowed
his men to be ambushed by a picked body of Franks, concealed
in a Roman amphitheatre :
So the army was scattered : but Folchere, left behind with his body
guard, would not deign thus to take to flight... He made stand as best he
could, with his back to a tomb, and slew many of the enemy, now rushing
upon them, now drawing back gradually with his face to the foe. And
while he might still have escaped easily, and his followers were urging
him to do so, "How," said he, "could I bear the tongue of Narses blaming
me for my recklessness? "...So he stood his ground and held out to the
last, till he fell forward upon his shield, his breast pierced with many
darts, and his head broken by an axe. Upon his body his followers fell,
to a man, perhaps voluntarily, perhaps because they were cut off1.
The Roman or Greek is separated from his heroic age by
too many centuries. Court intrigues, theological councils,
miracles wrought by the corpses of saints, are of more in
terest to him. With the German it is different ; after he has
been converted to Christianity the feeling remains as strong
as ever. For it is not only in the secular literature, in the Old
English Chronicle, in Maldon, or in Saxo that we find it. It
fires equally the Christian poetry, so that whenever the motive
of loyalty is touched, the gospel stories of the Heliand, or the
legends of apostles, become for the moment war-songs. It is
thus that, when St Andrew's disciples are bidden to leave him,
as he goes on a desperate errand, they reply:
Whither shall we, then, go ? Outcast, dishonoured, when men reckon
who best did to his lord war-service, what time in the field of battle
hand and shield endured hardship, smitten by the sword2.
Guthhere.
It is easy then to understand why the story of the fall of
Gundahari and his men in battle against the Huns was of
interest not merely to the Burgundian, but to all his neigh
bours, till, as the centuries passed on, it became known from
end to end of Germania. Eight centuries after his fight
Gundahari was still remembered from Iceland to Austria, and
1 Agathias, i, 15.
2 Andreas, 405-414 : cf. Wiglaf's speech in Beowulf (ed. Holder), 2864-91 :
cf. also 2633-60.
Stories Ttnown to Widsifh: Burgundian heroes 61
three elaborate collections of the stories which had been woven
round his name have come down to us. In the Nibelungen
Lied we have the version which some Austrian poet or poets
put together, about the year 1200. The Saxon versions, as
current in Bremen or Minister1, were put together in the form
of a prose saga — the Thidreks saga — by a Norseman, writing
perhaps not much later than the Southern poet2. Later still,
a little after the middle of the thirteenth century, the Norse
lays which had been sung during centuries in Iceland, Norway,
and Greenland, and which are still in great part extant, were
paraphrased and codified in the Vglsunga saga.
In all three cases another story, originally quite distinct, is
blended with that of the Burgundian downfall : the tale of
Sigfrid the Volsung, which told how Sigfrid slew a dragon
and won a mighty hoard of gold, and a maiden to be his
bride, whom he wakened from an enchanted sleep : and how
the Niblungs, his treacherous comrades, robbed Sigfrid of bride,
and gold, and life. Gundahari the Burgundian comes to be
identified with the Niblung king : at what date we cannot
exactly tell; but apparently Widsith does not know this
version of the story. Gundahari's violent death thus becomes
a just retribution for his sin. And so " the best tale pity ever
wrought " was formed around the historic figure of the Rhine-
land chief: "the Great Story of the North, which should be to
all our race what the Tale of Troy was to the Greeks — to all
our race first, and afterwards, when the change of the world has
made our race nothing more than a name of what has been — a
story too — then should it be to those that come after us no less
than the Tale of Troy has been to us3."
And thus Gundahari has been sung by the poets of fourteen
centuries : but on the other hand the historians have treated
him but shabbily. Gibbon ignores him, except for the mention
of his name in the Jovinus business : and so also do more
recent writers. Yet the overthrow of Gundahari is mentioned,
1 Thidreks saga, cap. 394.
2 "Sagaen er rimeligviis bleven til under Kong Haakon Haakonssons Eeg-
jering." — C. R. Unger, in his introduction to the Thidreks saga, p. iv.
3 WiHiam Morris : Introduction to his version of the Vglsunga saga.
62 Widsiih
briefly, by two contemporaries and by several later historians :
and the scanty, but quite reliable information which we thus
have from a number of Latin sources was pieced together by
Miillenhoff fifty years ago1. Combining the accounts of Prosper
of Aquitaine, and of Hydatius, Bishop of Chaves2, we have this
result. That in 435 or 436, a quarter of a century, that is,
after they had gained their footing in and around Worms,
Gundahari and his Burgundians again became restive, and
tried to press into Belgic Gaul : that they were met by the
Roman general Aetius, overwhelmed, and forced to beg for
peace. That within two years of this defeat Gundahari had
to face an onslaught of the Huns, who slew him, cum populo
atque stirpe sua to the number of twenty thousand.
Whether the later saga is true to history in representing the
Hunnish foemen as having been led by Attila, we cannot tell.
Paul the Deacon says it was so : but three centuries of heroic
song separate him from the historic battle. Paul is obviously
following an inaccurate popular account when he connects this
battle with the great invasion of Gaul twenty years later, which
resulted in Attila's repulse upon the Mauriac plain3. Elsewhere
1 Zur Geschichte der Nibelungensage in Z.f.d.A. x, 148-9.
2 t^36] Burgundiones, qui rebellaverant, a Romanis duce Aetio debellantur.
[437] Burgundionum ccesa xx millia. Hydatii Continuatio Chronicorum in
Mommsen's Chronica Minora, vol. n, pp. 22, 23 (M.G.H.).
[435] Eodem tempore Gundicarium Burgundionum regem intra Gallias habi-
tantem Aetius bello obtinuit [read obtrivii] pacemque ei supplicanti dedit: qua
non diu potitus est ; siquidem ilium Hunni cum populo atque stirpe sua
deleverunt. Prosper Aquitanus, in Migne, LI (1846), 596-7.
435. Gundicharium Burgundionum rer/em Aetius bello subegit, pacemque ei
reddidit supplicanti, quern non multo post Hunni peremerunt. Cassiodori
Chronica in Mommsen's Chronica Minora, n, 156 (M.G.H.).
It is Sidonius Apollinaris who tells us that the occasion was an incursion
into Belgica, in his Panegyric addressed to Avitus, c. 455 and therefore twenty
years after the event.
Aetium interea, Scythico quia scepe duello est
edoctus, sequeris; qui, quamquam celsus in armis,
nil sine te gessit, cum plurima tute sine illo.
nam post luthungos et Norica bella subacto
victor Vindelico Belgam, Burgundio quern trux
presserat, absolvit iunctus tibi.
Sidonii Carminum vn, 230-235. Ed. Luetjohann in M.G.H.
8 Historia Miscella, recens. Franc. Eyssenhardt, Berolini, 1869, pp. 323,
332, His etiam temporibus Gundicarium Burgundionum regem intra Gallias
habitantem Aetius patricius bello optrivit pacemque ei supplicanti concessit (Bk
xrv, cap. 11). Attila itaque primo impetu max ut Gallias introgressus est,
Gundicarium, Burgundionum regem, sibi occurrentem protrivit (xv, 4).
Stories known to Widsifh: Burgundian heroes 63
he more accurately distinguishes between the two events1.
Paul's account is chiefly to be valued as a witness to the
tradition of his day : Attila does not fall upon the Bur-
gundians in their town of Worms ; Gundahari marches east
ward to meet his death, " postquam sibi occurrentem Attila
protriverat." This has ever been an essential feature of the
story, whether told in High German, Icelandic, or English :
We will go and look on Atli, though the Gods and the Goths forbid;
Nought worse than death meseemeth on the Niblungs' path is hid,
And this shall the high Gods see to, but I to the Niblung name,
And the day of deeds to accomplish, and the gathering-in of fame2.
The statement that Gundahari was cut off with his race and
people is an exaggeration. A remnant was left, and some eight
or nine years after the disaster was settled in a new home in
Savoy. Burgundians played their part in the great battle
against Attila3. The nucleus (dwelling between the modern
Geneva and the modern Lyon) gradually extended their borders
till they again became a mighty nation. It was in these days
that, with their abundant animal spirits, their astonishing
appetites, and their more astonishing thirst, they vexed
Sidonius Apollinaris. The Burgundian code of laws, drawn
up before 516, under King Gundobad, contains a valuable
reference to " our ancestors of royal memory, Gibica, Gundo-
mar, Gislahari, Gundahari, our father and our uncle4."
1 Attila rex Hunorum, omnibus belluis crudelior, habens multas barbaras
nationes suo subjectas dominio, postquam Gundigarium, Burgundionum regem,
sibi occurrentem protriverat, ad universas deprimendas Gallias suce scevitite
relaxavit habena*.
De episcopis Mettensibus, in Pertz (fol.), SS. t. n (1829), p. 262.
2 William Morris — Story of Sigurd the Volsung.
* Jordanes, ed. Mommsen, xxxvi (191).
4 Si quos apud regia memories auctores nostros, id est: Gibicam, Gundo-
marem [or Godomarem], Gislaharium, Gundaharium, patrem quoque nostrum
et patruum liberos liberasve fuisse constiterit, in eadem libertate permaneant.
Lex Burg. p. 43, ed. L. B. de Salis, in M.G.H. Leges i, ii, 1. The correct
rendering of auctores has been disputed. Binding and Waitz translate " pre
decessors." Gundobad, they think, was not a descendant of the old reigning
house. Jahn on the other hand (pp. 300, 301, footnote) argues pretty con
clusively that auctores here must mean ancestors. Miillenhoff and Bluhme
would make the new race of kings descendants of the old, and the fact that
their names are compounded out of the same elements tells in favour of this
(Z.f.d.A. x, 153-4; vn, 527).
64 Widsith
Gifica and Oislhere.
The Gibica who thus stands at the head of the Burgundian
line is clearly the Gifica of our poem, the Gibeche of High
German, the Giuki of Scandinavian story. Whether he ever
had any historic existence is uncertain. The name, for the
founder of a dynasty, is suspicious, in view of the frequency
of the terms beaggifa, goldgifa, mafiftumgifa as synonyms for
" King1." No argument can be drawn from the occurrence of
this name along with those of historical kings, in the Bur
gundian laws : the Burgundian of the sixth century had, of
course, no doubts as to the historic existence of the ancestor
of his kings. In any case Gifica's reign, whether actual or
mythical, must be dated back to the old days when the
Burgundians dwelt in the East, and were neighbours of the
Ostrogoths in the forests of the Vistula. After the Burgundians
were settled in the West there would be a natural tendency
either to make him a king of the Burgundians in their new
home; or, keeping him in the East, to forget his connection with
the Burgundians altogether, and to make him a Hun or Goth.
And, in different branches of the old tradition, we find both
these things happening. In the Norse versions Giuki is father of
Gunnar; and sometimes too in the High German stories Gibeche
is the father of Gunter, and rules at Worms2. But in the
Nibelungen Lied, Dankrat8 is Gunter's father, while Gibeche is
a champion of the Hunnish court4.
Now it is noteworthy that when, in Widsith, Gifica is men
tioned as ruling the Burgundians, it is in immediate connection
with ^Etla and the Huns, with Eormanric the Goth and Becca
his liegeman, with the Kaiser who rules the Greeks, and with
1 Wackernagel, Sprachdenkmaler der Burgunden, p. 389 (in C. Binding).
Grimm (Z.f.d.A. i, 572-5) argues that the name, mythical in origin, ultimately
came to signify a Giver half or entirely divine.
2 As, for example, in the Rosegarden at Worms. In the Waltarius also,
Gibicho is father of Guntharius.
3 Grimm pointed out that Dankrat is possibly originally synonymous with
Gibica (Z.f.d.A. n, 573). See also Jahn, 126.
4 Der Nibelunge Not, ed. Bartsch, st. 1343, 1880. Compare Dietrich* Flucht,
7114, where Gibeche is a counsellor of Ermrich.
Stories knowi to Widsith: Burgundian heroes 65
C*lic of the Finns. The poet who drew up these lines had
learnt to think of Gifica as an Eastland hero : he probably
knew lays which represented Burgundian and Goth as neigh
bours : he felt no difficulty at any rate in grouping them
together1.
The Gislhere who is mentioned subsequently in our poem
with other heroes, presumably Gothic, is almost certainly to be
identified with the King Gislaharius of the Burgundian laws.
Later German tradition represents this prince as the chivalrous
younger brother of Gunther, Gisdher daz Joint*. But his place
in the list, as given in the Burgundian laws, would make him
a predecessor or ancestor ; and this is borne out by the way
in which his name occurs in Widsith. Ruling over the Bur-
gundians whilst they were still close neighbours of the Goths,
having dealings, whether in peace or war, with Gothic kings
and champions, he has been drawn into the Gothic cycle exactly
as has the Gibeche of Dietrichs Flucht, and like him is thought
of as a retainer of Ermanaric, or at any rate as a hero of the
Gothic cycle. That he was ever a Burgundian has apparently
been forgotten.
1 See note to 1. 19.
2 So in the Thidreks saga he is a younger brother (cap. 169). The part
played by Giselher in the Nibelungen Lied and Thidreks saga may, of course,
be due not to continuous tradition but to subsequent learned influence.
See Wilmanns Untergang der Nibelunge, 1903, pp. 23-4. Yet this seems
improbable.
C.
CHAPTEE III.
THE STORIES KNOWN TO WIDSITH. TALES OF THE SEA-
FOLK, OF THE FRANKS AND OF THE LOMBARDS.
The North-Sea tribes in the time of Tacitits. The Frisians and
Finn.
HITHERTO we have been following the stories of those chiefs,
Ermanaric, Attila, Gundahari, who led their German hordes
within ken of the historians of the Roman Empire, and con
cerning whose historical existence we therefore have definite, if
scanty, information. To turn from these stories to those of the
North-Sea folk, the native stuff of our English heroic poets, is
like moving from a room lighted by the occasional flash of a
dying fire to one where we have to grope in absolute darkness.
For to the Romans these Northern tribes were little known, and
that chiefly during the struggle under Tiberius and Germanicus.
It is one great merit of Tacitus' Germania that it preserves to
us fragments of that information concerning our forefathers,
which was acquired during this struggle. For by the time of
Tacitus himself the Romans had withdrawn west of the Zuider
Zee and of the Rhine : and these coast tribes had passed out of
their sight1.
The name Ingaevones which these coast tribes possessed in
common rested, Tacitus tells us, upon a belief that they were
all sprung from one eponymous ancestor Ing2. Most westerly
1 Exactly to what extent the Romans came into contact with the Frisians
is not certain. That the Frisian fisheries, on what is now the north coast of
Holland, were at one time worked by Roman publicans has been proved by an
inscription found near Leeuwarden. See Mnemosyne, xvi, 439, De inscriptione
romana apud Frisios reperta, by U. Ph. Boissevain. There were Frisian contin
gents, too, in the Roman army.
3 Germania, cap. n.
Stories known to Widsith : Tales of the Sea-folk 67
of all these maritime peoples dwelt the Frisians, amid those
'vast lakes navigated by Roman fleets1,' which have since become
the Zuider Zee. Tacitus recognises two groups of Frisians —
the Lesser Frisians, dwelling apparently to the west of these
lakes, near the Roman frontier, and the Greater Frisians,
dwelling in the unexplored marshes of the North and East.
Five or six centuries later, the same distinction seems to be
recognised by the poets of Beowulf and Widsith2. There are
the Western people, always associated with the Franks, against
whom Hygelac's raid was directed : and the main body of the
Frisians, whose traditional hero is Finn Folcwalding.
The tale of Finn must have been one of the best known
stories in seventh century England : of our five extant heroic
poems, or fragments of poems, three refer to it. The task of
harmonizing these references into one story is not an easy one.
But to investigate the Finn story in detail would take us too
far afield. The object of the present study is to see what we
can discover from Widsith about the lost Old English lays ; the
lay of Finnsburg is fortunately not all lost, and its reconstruction
demands a separate study3. The popularity of the story in
England is not to be attributed to Frisian immigrants4. In the
great fight in Finn's hall, most of the North-Sea tribes seem to
have claimed their share, and the tale was probably common at
an early date to them all, and was brought across the North
Sea by gleemen of Saxon and of Jutish race5.
Chauci and Saxons.
The Frisians were divided from their next neighbours, the
Chauci, by much the same boundary as now separates the north
coast of Holland from that of Germany. All the space up to
1 Cap. xxxiv. 2 see note to 1. 68.
8 References will be found in Brandl, in Pauls Grdr.(2) n, i, 986 and in Heyne-
Schiicking, 112. For Finn Folcwalding see also Bieger, Z.f.d.A. xi, 200 ; Grimm,
DM. 218; transl. Stallybrass, i. 219.
4 'Bpirrlav W ri]v vrjffov (Ovij rpla iroXvavOpwirbraTa i;X0vo'l----'Ayyi\oi re Kal
iffffoves Kal ol rrj vr)<T(f 6/auvvfjiOi 'Bpirruves. Procopius, Bell. Gott. IV, 20.
But this statement as to Frisian immigrants lacks confirmation, and is
probably due to a confusion of Frisians and Saxons.
5 Binz in P.B.B. xx, 179-186. Binz points out that men and places bearing
the names of heroes of the Finn story are found chiefly in Essex, and in that
part of Wessex which had originally a "Jutish" population: but not, oddly
enough, -in Kent.
5—2
68 Widsith
the Elbe was occupied by the Chauci, of whom Tacitus draws
an idyllic picture. Populus inter Germanos nobilissimuSy quique
magnitudinem suam malit justitia tueri. Sine cupiditate, sine
impotentia quieti secretique nulla provocant bella\ Beyond the
Elbe, in Holstein, dwelt the Saxons, who, though they are
strangely enough passed over in silence by Tacitus, are located
by Ptolemy's reference to them as dwelling " upon the neck of
the Cimbrian Peninsula."
By the fourth century the name of the Chauci has dis
appeared, and the whole district from Holstein to the Frisian
border is apparently filled with people calling themselves
Saxons2. Clearly some confederation, whether friendly or
forcible, has taken place, by which the Saxon name has come
to be applied to both people. The Saxons who colonized
England include, then, the Chauci of Tacitus, in addition to
the little tribe in Holstein which originally bore the Saxon
name3, but which clearly would not have sufficed to settle
Essex, Middlesex, Sussex, Wessex, and the lands by the
Severn.
The main body of the Saxons remained in Germany, but by
constant conquest and assimilation of Frankish and Thuringian
tribes, they became a great people, and at the same time lost
their distinctively Anglo-Frisian character. Consequently the
modern dialects of North- West Germany are not "Anglo-
1 Germania, xxxv. The Chauci are classed by Pliny among the Ingaevones,
Nat. Hist, iv, 14 (28) "Ingyaeones quorum pars... Chaucorum gentes."
2 Probably the change took place in the third century, when the Saxon
name again meets us for the first time since Ptolemy, whilst the name
Chauci disappears (see Weiland, 26-7). Hence there is no mention of the Chauci
in Widsith, unless we are to accept the exceedingly doubtful conjecture of
Miillenhoff (see note to 1. 63). Grimm (G.D.S., 675, on which see Rieger,l£./.d..4.
xi, 186-7) would see possible survivals of the name Chauci in the Hugos of
Beowulf and the Hocingas of Widsith. In the latter conjecture Grimm had
been anticipated by Miillenhoff, Nordalb. Stud, i, 157 ; in the former by Ett-
miillerd) 16. Neither conjecture seems likely. It has been urged in support
that the district near the modern Groningen was of old called Hugmerke, which
has been interpreted "the border of the Hugas " (Chauci). But the support to
be derived from this is weak geographically, for there is no evidence that the
Chauci ever extended so far west; and weak etymologically, for other inter
pretations of Hugmerke need to be considered. See Bichthofen, Untersuchungen
ttber Friesische Rechtsgeschichte, Berlin, 1882, Th. n, Bd 2, pp. 754-5; and
cf. Weiland, 29. That Hugas is equivalent to Chauci has been maintained by
Much (94) ; for reasons to the contrary cf. Loewe in I. F., Beiblatt, xiv, 22.
3 Moller, V.E. 84; Weiland, 31.
Stories known to Widsith : Tales of the Sea-folk 69
Frisian1/' and the Heliand, or the Old Saxon Genesis, though
they perhaps show " Anglo-Frisian " peculiarities, are not
written in an " Anglo-Frisian " dialect. These poems, though
Christian in subject matter, are in style and spirit closely akin
to the old heroic poetry2: whilst the old tales of Ermanaric and
Gundahari, Weland and Wade, must have lived on side by side
with the religious poetry, till they were woven by a Norwegian
of the 13th century into a more or less continuous story around
the figure of Theodoric of Verona. This Thidreks saga, Norse
in diction but Saxon in origin, gives as we have seen the most
valuable sidelights upon the English, as well as upon the Norse
and High German versions of the old tales.
The Worshippers of Nerthus.
East and north of the Saxons and Chauci dwelt, in the days
of Tacitus, a confederation of tribes3. Unfortunately, no common
name has been preserved, but we know from Tacitus that the
Reudigni, Aviones, Angli, Varini, Eudoses, Suardones, Nuithones,
dwelt among fastnesses of wood and water and worshipped in
common Nerthus — that is, Mother Earth — believing her to
intervene in human affairs.
There is a sacred grove in an island of the ocean, and in the grove a
car, dedicated and veiled. One priest alone has access to it. He can
tell when the goddess is present, and he attends her as she is drawn by
a team of cows in reverent procession. Then there is holiday wherever
the goddess condescends to visit as a guest : they do not go forth to
war ; they do not take up arms ; all weapons are laid aside : then and
then only peace is known and loved, till the same priest brings back the
1 See Bremer in Pauls Grdr. (2> in, 850, 860-73.
2 Nowhere is the loyalty of the gesith to his lord shown better than in the
relation between Satan and his followers in the Old Saxon Genesis, as repre
sented in the so-called Ccedmon, or the relation of Thomas to his Lord, as shown
in the Heliand (3993-4002). Cf. Olrik, Heltedigtning, 81.
8 Tacitus mentions these peoples among the Suevic tribes, who often have
been considered as more or less equivalent to the "inland" Herminones. Yet
these seven tribes certainly dwelt proximi Oceano, and should therefore rather
be classed among the Ingaevones. They were certainly "Anglo-Frisian."
Tacitus has been enumerating the inland people, and as he goes north he
continues to enumerate among the Suevic peoples first the Lombards, then the
Angles (note, however, that Widsith distinguishes Engle and Swsefe whilst
grouping them together). But Tacitus admits ignorance : Et hac quidem pars
Suevorum in secretiora Germanice porrigitur. Propior, ut . . . Danubium sequar,
Hermundurorum civitas. Accordingly he proceeds to trace the Suevi along the
more familiar border lands of the Danube.
70 Widsith
goddess to her grove, tired of human converse. Car, draperies and, if
you will believe it, the very divinity are washed in a secret lake, in which
lake the slaves who do the work are forthwith drowned. Hence a mystic
terror and a pious ignorance of what that may be which is seen only by
those destined to die.
Strange has been the history of this goddess Nerthus in
modern times. Sixteenth century scholars1 found irresistible
the temptation to emend the name of "Mother Earth" into
Herthum, which nineteenth century scholars further improved
into Hertham, Ertham, For many years this false goddess
drove out the rightful deity from the fortieth chapter of the
Germania*. Antiquaries identified the " island of the ocean "
with Rtigen, which happened to possess a convenient lake, and
diligent enquiries as to Hertha in that locality succeeded at
last in inoculating the inhabitants with a tradition regarding
her. According to one assertion, she even found her way into
the proverbs of the Pomeranian peasantry3. And so an early
Teutonic goddess Hertha was turned loose into the field of
comparative mythology. Without wishing to trespass on that
field, or rather
Serbonian bog
Where armies whole have sunk,
it is enough to state that there can be no doubt that the correct
reading in the Germania is Nerthus, and that no goddess Hertha
was ever worshipped by the ancient Angles. The Scandinavian
form corresponding to Nerthus would be NJQrthr, and such a
god4 was worshipped with his son Ingvifreyr at old Uppsala,
with rites resembling those which Tacitus describes as rendered
1 Perhaps we might say fifteenth century scribes: for some transcripts of
the Germania of that period read Herthum, and one even nisi quod mam-
menerthu (see Germania, ed. Massmann, 1847, pp. 18, 118). It must have
been upon this reading that Holder based his Mammum Erthum— Mother Earth
(Germania, ed. Holder, 1882).
2 Amongst editors who have bowed the knee to Hertha may be mentioned
Oberlin (1801), Bekker (1825), Walther (1833), Ritter (1834, 1836), Kritz (1860,
Ertham). The Aldine edition of 1534 reads Herthum, following the reading of
Beatus Rhenanus in his edition of the preceding year.
9 Grimm, trans. Stallybrass, i, 256 ; Miillenhoff, D.A., iv, 470-1.
4 The change in sex would be rendered easier by the fact that the u-stems
were always, in the main, masculine, and in Norse became exclusively so. A
stage at which a male and a female Nerthus were venerated side by side seems
to be indicated by Heimskringla, i, 4 ; Lolcasenna, 36. For Njorth, see Mogk in
Pauls Grdr.(2) in, 323 etc., 367 etc.; R. Meyer, 204 etc.; Golther, 218 etc., 456 etc.
For the change from Feminine to Masculine see Kock in Z.f.d.A. xxvin, 289,
Historisk Tidskrift (Stockholm) 1895, 166, note.
Stories known to Widsith: Tales of the Sea-folk 71
to Nerthus. The coupling of Ingvifreyr with Njorth suggests
the question whether Ing — otherwise known as Frea, the lord —
may not have been coupled with Nerthus in the worship upon
the isle of the ocean.
The island centre of this Nerthus- worship we might at first
sight be tempted to identify with Heligoland — the "Holy Island"
of the North Sea1. But that little rock, with its grass-crowned
red cliffs, and its stretches of white sand, could never have
given root to the sacred grove of the goddess2. We must
rather place this in one of the (now Danish) islands near the
Great Belt: perhaps in Zealand3, from which the spreading of
the cult across the Sound to the Scandinavian lands can have
most easily taken place. It is highly probable, at any rate, that
the sanctuary was situated in one of the islands of the Baltic.
For there is hardly room to locate the seven tribes of Tacitus all
upon the mainland ; some of them must have occupied the
adjacent Baltic islands.
\
The limits of Angel.
Off a weold Ongle...
ac Offa geslog cerest monna
cnihtwesende cynerica meest.
The best of all possible evidence that the Angles and their
kindred tribes once held the islands adjoining is given by a
passage added to the old English translation of Orosius, in
which Ohthere describes to his lord, King Alfred, his voyage
from Sdringesheal (in the Bay of Christiania) to wt Haethum
1 The identity of Heligoland and of the island sacred to Nerthus has been
maintained by Schade (Die Sage von der heiligen Ursula, Hannover, 1854, pp.
113-118). Beyond the fact of Heligoland having been in later times a sacred
island, and having possessed a sacred spring, there is little to be said in favour
of the identification, and much against it. Tacitus' description seems to
demand either an island large enough for the processions of the goddess, or
sufficiently near to the mainland to allow the chariot and its team of cattle to
be ferried across. Of. also Miillenhoff in Nordalb. Stud, i, 128. Weiland in
1889 thought it "no longer open to doubt that the sacred island of Nerthus was
in the North Sea, not in the Baltic " (Die Angeln, 9). Becent criticism has
tended to reverse this judgement.
2 habet arborem nullam. Adam of Bremen, rv, 3, in Pertz (fol.) SS. vn,
p. 369.
3 Much in P.B.B. xvn, 195-8.
72 Widsith
(Haddeby, the modern Schleswig). The voyage takes five
days. For the first three he sails south down the Cattegatt.
"Then was Denmark" [the original home of the Danes in
Southern Sweden, which till many years later was an essential
part of Denmark] "Then was Denmark to his left, and the
open sea to his right, three days." [Here Ohthere reaches the
Sound.] "And then, two days before he came to Haddeby,
was Jutland to his right, and Sillende [? Zealand1], and many
islands. In those lands dwelt the Angles, before they came
hither to this land." [Having passed the Sound, Ohthere has
to turn to the right, and steers his way amid the islands,
keeping Zealand (and Jutland) still on his right, but having to
his left the smaller islands Falster and Laaland, which are
mentioned just below as belonging to the Danes.] " And then
for two days he had on his left the islands which belong to
Denmark."
The importance of this passage can hardly be exaggerated.
King Alfred, we have seen, was deeply versed in old English
poetry, and must have known dozens of lays, lost to us, dealing
with the old kings of Angel, their tributary kings, their feuds
and alliances with other Baltic- Sea folk. And here we have
an authority which we can hardly regard (in this part, at any
rate, of the Orosius) as other than that of Alfred himself,
making not merely Jutland, but also many islands, part of the
original Anglian home.
The whole of this territory was probably, in the first century,
occupied by the Nerthus-worshipping people. Exactly how we
locate or identify the seven tribes mentioned by Tacitus does
not really much matter. Tacitus has been enumerating tribes
from south to north — first the Semnones, then the Langobardi —
1 Sillende is probably Zealand (O.N. Selund, or Silund, called sometimes Siel-
lendia, Sjalendia, in Latin-Danish documents). The difficulties which have
been raised with respect to this passage are quite gratuitous. It is unnecessary
to apologize for taking the " Denmark " on Ohthere's left as Southern Sweden,
by the plea that " Southern Sweden was then in the hands of Denmark "
(M.L.E. iv, 273). This district was then as much a part of Denmark as Jut
land is now : and Alfred, like Adam of Bremen, calls it by its name. The Sweden
of this date was a most restricted area. The difficulty arises only if we insist
upon understanding Alfred's terms in their twentieth century meaning. The
alteration of the text suggested by Schuck (Folknamnet Geatas, p. 19) is un
necessary. See Appendix L, The geography of the Orosius compared with that
of Widsith.
Stories known to Widsith : Tales of t?ie Sea-folk 73
and the use of the word deinde implies that he is still enumerat
ing them in this order : and so far as we can check him we find
that he is. The Reudigni, then, would be most to the south, in
Holstein. But on this " neck of the Peninsula " Ptolemy placed
the Saxons. One critic therefore supposes " Reudigni " to be a
nickname of the Saxons1. The arguments in favour of this
identification are exceedingly inconclusive, but there is nothing
to be said against it ; and as it enables us to find a home for
the Reudigni, who would otherwise be destitute, and at the
same time accounts for the extraordinary omission of the
Saxons by Tacitus, it seems convenient to accept it as a
possible hypothesis1. Then come, in Tacitus' enumeration, the
Aviones and the Angli. The modern Angeln enables us to
locate the Angli proper in and around southern Schleswig:
the Aviones, as neighbours both of the Saxons and of the
Angles, are most easily placed in Fiinen and the other islands.
Following a hint of J. Grimm2, Much3 would interpret the name
with great probability as the " island folk," aud this interpreta
tion supports the theory of their home in the Baltic Isles. The
Aviones are, in all probability4, to be identified with the Eowan,
who are mentioned, together with a number of these sea-folk,
and their heroes, in Widsith.
Breoca [weold] Brondiugum, Billing Wernum ;
Oswine weold Eowum, and Ytum Gefwulf,
Fin Folcwalding Fresna cynne.
Sigehere lengest Sse-Denum weold.
The Varini are certainly4 the Wernas of Widsith : following
Tacitus, we should expect to find them just north of the Angles
in northern Schleswig : and traces of their name are still to be
found there. With less certainty, though still with some proba
bility4, we can identify the Eudoses with the Ytas of our poem
and with the Jutae, who, according to Bede, colonized Kent
and the Isle of Wight. Following Tacitus, we should put them
1 Much in P.B.B. xvii, 191-5. Miillenhoff hazarded the guess that the
name Reudigni was that of a small people subsequently merged in the more
comprehensive Saxons (Nordalb. Stud, i, 119). See also Weiland, p. 27.
2 G.D.S. (first ed.) 472, (fourth ed.) 330. But Miillenhoff had anticipated
Grimm, Nordalb. Stud, i, 118 (1844).
3 P.B.B. xvn, 195.
* See the footnotes to Wernum, Eowum (11. 25, 26) and note on Ytum (Ap
pendix D, The Jutes).
74 Widsiih
in southern Jutland, the more so as this agrees with Bede's
information about the Jutes. There is still territory left in the
north of Jutland in which to settle the Suardones [or Suarines]
and Nuithones — the former conceivably identical with the
Sweordweras of our poem1.
Most, then, of what is now Denmark and Schleswig-
Holstein was in the days of Tacitus held by the Nerthus-tribes,
with the Angles dwelling in the centre, and probably, at any
rate at a later date, giving their name to the whole confedera
tion. For it is not likely that Britain, from the Forth to the
Stour, and the Wash to the Severn, was conquered and settled
by the little central people alone : and, though we cannot trace
in English place-names any of the minor tribes except the
Varini and the Jutes, the evidence of Alfred, who brings the
Engle from Jutland and the islands, shows that he at least
believed that all joined in the settlement of England. It
is of particular importance to the study of Widsith to observe
clearly what Alfred says. For several eminent scholars2, as
suming that King Alfred and Bede state that the Angles came
from Angel, and that the ancient Angel must necessarily have
been coterminous with the modern Angeln, have argued : that
Angeln cannot have been the home of the ancient Angles,
because it is too small to have held them all : that therefore
King Alfred and Bede are wrong, and that the home of the
Angles was somewhere far inland, on the banks of the middle
and upper Elbe. This view cannot be passed over in silence
by a student of Widsith, since our poem definitely localizes
Offa, King of the Angles, in Angel. If, then, the connection
between the Angles and Angeln rests, as Zeuss thought, upon
a piece of false etymology of Bede's time or a little earlier, we
must not attribute Widsith, in its present form, to a period
much earlier than Bede. The assumption that the ancient Angel
was no larger than the modern Angeln is, however, a gratuitous
one : on the contrary, we have only to think of Cumberland,
Northumberland, Lombardy, Burgundy, Servia to see that when
a national name remains, while the nation, as such, is passing
1 Although the two names do not correspond phonetically. See note to 1. 62.
a E.g. Zeuss, 496; Steenstrup in Danmarks Riges Historie, i, 76-77.
Stories known to Widsith : Tales of the Sea-folk 75
away, the waning of the nation is naturally accompanied by a
shrinkage in the application of the name. Alfred's statement
as to the wide extent of the territory occupied by the Angles
before their emigration to Britain may be taken as indicating
that here, as everywhere else in Germany, a confederation of
the smaller tribes of Tacitus' day had taken place, and that the
name Angli had come to be applicable to the whole or a great
portion of the Nerthus- worshipping tribes.
The Danes.
The immediate neighbours of the Angles (using the term in
the broad sense) were the Danes, dwelling originally on what is
now the southern coast of Sweden1. These two seafaring people,
separated only by the narrow seas, must have been em
phatically either friends or enemies. Especially must this
have been the case after the Angles began to migrate to
Britain and the Danes began to settle in the vacant lands —
two events which followed hard upon one another. The emi
gration of the Angles, we know, lasted through the latter half
of the fifth century and great part of the sixth. Already by the
beginning of the sixth century, Danes had probably occupied
most of Jutland as well as Zealand2. By the end of the sixth
century, it is asserted, they were in possession up to the Eider3.
Now this Danish settlement may have been a hostile one ; but,
on the other hand, it may have taken place by a friendly ar
rangement, an understanding that Danish immigrants might
claim any vacant land, but must not disturb such of the old
inhabitants as chose to remain behind. A friendly settlement
1 It must be noted that although Tacitus, Pliny and Ptolemy know the
Cimbric Chersonnese, and Ptolemy knows it well, none of them know of the
Danes, unless the Danes are to be identified with the Aavduvts of Ptolemy
u, 11, 16. The Daukiones are neighbours of the Goutai in the south of
Skandia. Jordanes (cap. in, 23) mentions the Danes, and the context would
lead us to suppose that he placed them in Scandinavia. From a passage in
Procopius (Bell. Gott. n, 15) we should gather that there were Danes in
Jutland about the year 510, and from this time onward mention of the
Danes is fairly frequent. As the Angles pass into Britain, the Danes, stepping
into their places, come within the view of the Roman and Greek writers.
2 See Holier V.E. 5 (footnote).
3 Bremer in Pauls Grdr.(2) in, 837.
76 Widsifh
seems quite likely on a priori grounds. It is true that we hear
little of such arrangements during these centuries, because the
historians are only concerned with territorial squabbles just
when they lead to big battles and wholesale slaughter; yet
the constant migrations would have been impossible without
some mutual understandings. The battles we hear of spring
from those cases where rival claims could not be harmonized1.
Beowulf shows that the northern courts of the sixth or seventh
century were not ignorant of diplomacy. And in the case we
are considering there would be no cause for quarrel. Danish
settlers would find enough vacant land, and more than enough,
for Angel proper remained desert till Bede's time. The waning
Anglian folk, deprived of their best warriors, and left with more
old men, women, children and cattle than they could well
protect, would live in fear of sea-raids of Vikings, whether
Heruli or Heathobards, or land raids from the Suevic folk to
the south, who had ancient grudges to pay off dating from the
time of Offa. The remaining Angles would see with satisfac
tion stalwart Danes coming, not to plunder, but to settle and to
take up land in neighbourhoods which had been left naked and
defenceless2. The newcomers would soon, and all the more
quickly if they started with a common ancestral hatred for
Heathobard plunderers, be united to the old inhabitants by
their common interest in defending the country. There is a
good deal of circumstantial evidence that things happened in
this way.
Saxo Grammaticus begins his Danish history with the
brothers Dan and Angul8. He describes them as the joint
founders and rulers of the Danish race : from Dan sprang the
royal line of Denmark ; whilst Angul gave his name to the
1 Cf. Gregory of Tours, Historia Francorum, v, 15. A quarrel arises from
the unexpected return of a tribe to territories which they had vacated, and
which they find occupied by new settlers. Negociations take place, which
break down owing to the excessive claims of the original owners. Definite
agreements between tribes who invaded new lands and those who occupied their
old homes are mentioned by Procopius, Bell. Vand. i, and Paul the Deacon, n, 7.
2 The Danes in Widsith and Beowulf are not a sea-roving people, but are
rather busied in defending their own borders from pirates.
3 Dan igitur et Angul, a quibus Danorum cepit origo . . . non solum conditores
gentis nostre, verum eciam rectores fuere (Saxo, i, 1). See also Bremer in Pauls
in, 853.
Stories known to Widsith : Tales of the Sea-folk 77
province he ruled, and his successors, when they obtained
possession of Britain, changed its name to a new one derived
from their fatherland. Now the Danes must have been quite
aware that the English, with their widely-differing dialect, were
not as near of kin to them as the Swedes, Geats, and Norwegians,
who spoke an almost identical tongue. The place of honour
given to Angul as one of the two sources of the Danish race
probably points back to a time when Danish bards and
genealogists were conscious that their nation contained an
Anglian element.
Again, we have seen that the cult of Ing and Nerthus spread
into the heart of Sweden. This cult can have passed from the
isle of the ocean mentioned by Tacitus to old Uppsala only
through the intermediate Danes, who still at this time held the
whole of what is now the south coast of Sweden ; and friendly
relations and inter-marriage rather than incessant warfare
would make it easy for the Ingaevonian worship to spread
among the Danes, till we find in Beowulf an Anglian poet
bestowing upon them the title of "friends of Ing1." The
Angles came to regard the Danes as peculiarly the wor
shippers of Ing, for at last an English poet says "Ing was
first seen among the east Danish folk2."
The silence of our authorities is also strong evidence that
the Danish settlement of Anglian lands was a quiet and
peaceable one. Had there been a protracted struggle between
Angle and Dane, heroic defences of old halls against over
mastering odds, these things would probably have passed into
legend and song. But we hear nothing of such contests. On
the contrary, every reference to the Danes in old English verse
1 Hrothgar is twice called protector or prince of the Friends of Ing, eodor
Ingwina (1044), frea Ingwina (1320). His more usual title is frea (eodor)
Scildinga. It would be interesting to know whether an English poet of the
seventh century regarded Ingwine and Scildingas as synonyms, or if the
names are felt to refer to two sections of a mixed people, over whom the
Danish king rules. Cf. Sarrazin in Engl. Stud. XLII, 3, 11, etc.
Ing wees cerest mid East Denum
gesewen secgun, oj> he sffiftan est [eft Grein, Kluge]
ofer wag gewat; ween after ran;
~5w« Heardingas ftone hcele nemdun.
Runic Poem (67-70).
We must not however forget that Tacitus does not count the Nerthus-tribes
as Ingaevones, but as Suevi.
78 Widsith
is couched in the most friendly and respectful tone : a tone
which would be inconceivable if the English settlers had sailed
from the old Angle land leaving behind them burning home
steads, slaughtered friends, and triumphant Danes1.
And this gradual settlement of the Danes amid the
gradually departing Anglian tribes would explain why old
English poetry is interested in the Danes: their kings, their
foes, and above all their great Hall. It is not, as we might too
hastily assume from Beowulf, that the English were normally
more concerned with the doings of Scandinavian kings than
with those of their own heroes ; for Beowulf is exceptional in
the interest shown in the warfare of Swedes and Geats. Putting
together all the references we can find in English to heroic
stories, we find no particular interest in these matters2.
Geats and Swedes and the Swedish king Ongendtheow are
just mentioned in Widsith. But on the other hand in Widsith
we hear of four different Danish kings: Alewih, Sigehere,
Hrothwulf and Hrothgar; whilst the fight at Heorot must
have been very well known, for the names of the hostile
chiefs, Froda and Ingeld, are to be met with often enough in
England3. Alcuin the Northumbrian, writing to a bishop of
Lindisfarne, quotes Ingeld as the typical hero of song4. That
the English should have long felt an interest in the fate of
Heorot is, after all, natural, when we remember that the ground
on which it stood had perhaps been theirs, and may have been an
Anglian stronghold or sacred place. Bold theorists have even
1 The argument is stated very well by Miillenhoff, Beovulf, 88. But the
case is even stronger than Miillenhoff saw; for the lines in Widsith do not
point to any conflict between Offa and Alewih the Dane. It is merely stated
that Alewih, though the bravest of men, did not surpass Offa (see note to 1. 37).
The name of Alewih was remembered in Mercia, and comes into the pedigree
of Offa II. See also Weiland, p. 40; otherwise Binz in P.B.B. xx, 169.
2 The name Hygelac is met not infrequently in northern England.
3 See Binz in P.B.B. xx, 173-4.
4 Verba dei legantur in sacerdotali convivio; ibi decet lectorem audiri, non
citharistam, sermones patrum, non carmina gentilium. Quid Hinieldus cum Christo ?
The king of Heaven must not be mentioned with pagan kings who are now
lamenting in Hell. The passage is quoted by Oskar Janicke in his Zeugnisse
und Excurse zur deutschen Heldensage (Z.f.d.A. xv, 314), on the authority of a
MS of which he does not give particulars. The reference to Ingeld does not
occur in the received text (Letter to Speratus, ed. Froben, 1, 1, 77). It will be
found however in P. Jaffa's Monumenta Alcuiniana (Bibliotheca Eer. Germ, vi),
Berlin, 1873, p. 357 ; Epistolee, 81.
Stories known to Widsiih : Tales of the Sea-folk 79
identified the town where Hrothgar built Heorot1 with the
sacred place of the Nerthus-cult : and the holy grove with
its dread lake in Tacitus may remind us of the mere, over
shadowed by trees, in Beowulf.
The onslaught at Heorot.
Hro^vndf and Hroftgar...
forheowan cet Heorote Heafto-Beardna tyrym.
The chief value of the references to Heorot in Widsiih lies
in their correcting the impressions which we get from Beowulf.
Heorot is not thought of by our poet as the place where
Beowulf fights with monsters, but as the scene of a struggle
between Danes and Heathobards.
This struggle is not forgotten in Beowulf, but it is pushed
into the background, and appears as a mere ornament to
decorate the speech of the hero, and show his foresight.
Beowulf, on his return home, tells his king how the old feud
between Dane and Heathobard has for the moment been settled,
and how Freawaru, the daughter of Hrothgar, has been given
in marriage to Ingeld, the son of the Heathobard chief Froda,
who had fallen in the feud. But the peace, Beowulf thinks,
will not be lasting. Some survivor of the fight will urge on
the younger man to revenge, showing him a young Dane
wearing his father's sword :
Canst thou, my lord, tell the sword, the dear iron, which thy father
carried to the fight, when he bore helm for the last time, where the Danes
slew him and had the victory?.. .And now here the son of one of those
slayers paces the hall, proud of his arms, boasts of the slaughter and wears
the precious sword which thou by right shouldst wield.
Nothing could better show the disproportion of Beowulf,
which "puts the irrelevances in the centre and the serious
things on the outer edges2," than this passing allusion to the
story of Ingeld. For in this conflict between plighted troth
1 See Much in P.B.B. xvn, 196-8, Mogk in Pauls Grdr.($ in, 367, Kock in
Historisk Tidskrift (Stockholm) 1895, 162-3, identifying Leithra with the
Nerthus-temple ; and an article by Sarrazin, Die Hirsch-Halle, in Anglia, xix,
368-91 ; also his Neue Beowulfstudien and Der Grendelsee in Engl. Stud. XLII,
6-13, identifying both Leithra and the Nerthus-temple with Heorot.
2 Ker, Dark Ages, 253.
80 Widsith
and the duty of revenge we have a situation which the old
heroic poets loved, and would not have sold for a wilderness
of dragons.
Perhaps the speech of the old warrior in Beowulf is taken,
with little or no modification, from an Ingeld lay1. In any case,
this egging on is evidently an essential, perhaps the essential
part of the story, for it reappears in the Danish song, as para
phrased by Saxo Grammaticus. His version however is dis
torted. Frotho, the ancient foe of the Danes, has become a
Danish king2, slain by the Saxons. Yet the speech of the eald
cescwiga survived these changes, and has provided Saxo with
material for eighty sapphic stanzas.
lit quid, Ingelle, vicio sepultus
Vindicem patris remoraris ausum ?
Num pii cladem genitoris equo
Pectore duces 1
Re magis nulla cuperem beari
Si tui, Frotho, juguli nocentes
Debitas tanti sceleris viderein
Pendere penas3.
Needless to say the old warrior's egging on is successful.
Ingellus is stirred, and successfully avenges his father.
Saxo is correct, in so far as he preserves the tradition of a
Danish victory. For Widsith shows us that in the original
story the Danes were victorious ; but both Widsith and Beo
wulf show that in this story Ingeld was the leader, not of the
victorious Danes, but of their defeated foemen4. Probably
1 Miillenhoff, Beovulf, 43; Moller, V.E. 105.
2 Hence the degradation of Frothi in later Scandinavian tradition, where he
becomes a slayer of his kin (Saga of Rolf Kraki, Grdttasyngr). This is because
Frothi, having been taken into the line of Danish kings, could be regarded as
having a feud with the family of Halfdan only by the supposition of domestic
strife. Cf. Neckel, Studien iiber Frdfti in Z.f.d.A. XLVIII, 182, and Heusler, Zur
Skioldungendichtung in Z.f.d.A. XLVHI, 57.
3 Ed. Holder, pp. 205-13. For a parallel between the wicinga cynn of
Widsith and the viking life as outlined in the song translated by Saxo, cf.
Neckel in Z.f.d.A. XLVIII, 183 etc.
4 Koegel has argued that Saxo, in representing the attacking party as
Danes, is true to the original version. He thinks that in this original story
Hrothgar and his people were Angles of the islands, defending their homee,
still with success, against the forward pressing Danes (Heafto-beardan). Later,
hi the 7th or 8th century, when the fact that of old the Angles dwelt on
the islands of the Gattegatt had been forgotten, it was assumed that a king
ruling there must necessarily have been a Dane, and the story recast accord
ingly (Ltg. i, 1, 157). Not only is there much to be said against this inter
pretation, but also the grounds upon which it rests are quite unsound, for
Stories known to Widsifh : Tales of the Sea-folk 81
Ingeld1 was slain in this struggle, and certainly the eald cescwiga
fell, fighting to the last. In Scandinavian story the death of
this old warrior, Starkath, was remembered, though not in con
nection with Ingeld's revenge. According to one version his
severed head bit the earth when it met the ground : in another
story " the trunk fought on when the head was gone2."
The Lords of Heorot.
Hro^wulf and Hrofigar heoldon lengest
sibbe cetsomne suhtorfcedran.
It was possible for Saxo to transform the foes of the Danes,
Frotho and his son Ingellus, into Danish kings without deposing
their adversaries Rolf and Roe (Hrothwulf and Hrothgar) from
their true place as Danish monarchs. For he has divorced the
story of Ingellus' revenge entirely from that of Hrothgar and
his kin, which he gives in a quite different part of his book.
This Danish story of Hrothgar and Hrothwulf, though it
preserves the kinship exactly as in the old English tradition,
represents differently the parts played by the members of the
Danish house towards each other. It is preserved not only in
Saxo, but also in the Saga of Rolf Kraki, which was written
down from tradition some two centuries after Saxo's time.
Haldanus as Saxo calls him, Halfdan as he is in the Saga,
leaves two sons Roe (Hroarr) and Helgo (Helgi). Roe rules
Denmark and founds Roskild, but Helgo becomes a mighty sea
ting — he is, in fact, that Helgi Hundingsbane who is the hero
of the Eddie Helgi lays. In time Roe is slain — his slayer in
Saxo is called Hodbroddus, in which name we may perhaps see
(1) we have seen that even in the 9th century the fact that the Angles had
once held the islands of the Cattegatt was not forgotten, and (2) it is nowhere
explicitly stated in Beowulf or Widsith that Hrothgar did rule these islands.
We simply infer it from the fact that he and his people are called Scyldingas,
Dene. There are in fact no geographical data in Beowulf sufficient to enable
us to identify Hrothgar as a Dane : we only know that he is one because the
poet tells us so. Had Hrothgar been an Angle in the original story, there is,
then, nothing to have prevented him from having been localized in England,
as, in spite of much more serious difficulties, Offa was.
1 Sarrazin has suggested a connection between our Ingeld and another
Ingellus of Saxo (Book n, ed. Holder, p. 56). See Engl. Stud, xxin, 233. There
seems very little in favour of this.
Saxo vin (ed. Holder, p. 274) ; Helga kvfya Hundingsbana, n, 19. See also
Bugge, H.D. 157.
c. 6
82 Widsith
a survival of the Heathobeardan, the foes of Hrothgar in
Widsith1. But Helgo avenges his brother, and in course of
time is succeeded on the throne of Denmark by his son
Roluo (Hrolfr Kraki).
In Beowulf the same four Danish chiefs meet us. The
''high" Healfdene has sons Heorogar (not paralleled in Saxo),
Hrothgar and Halga the good2. But Hrothgar is not, as in
Danish tradition, slain in war, and avenged by Halga : on the
contrary, he outlives Halga, and we find him both in Beowulf
and Widsith with a nephew Hrothulf, who is evidently Halga's
son.
For a very long3 time did Hrothwulf and Hrothgar keep the peace
together.
These words, in Widsith, seem to imply that at length the
bond between Hrothwulf and Hrothgar was broken : and that
it was we gather also from the hints in Beowulf, where we see
Hrothulf always overshadowing Hrothgar's sons and nephew, —
those younger cousins who should stand nearer to the throne
than he.
In Beowulf 'it is to be noted that, although Halga is clearly
the younger brother, his orphan son Hrothulf is represented as
older than Hrethric and Hrothmund, the sons of king Hrothgar.
For Hrothulf has won his place of honour, and is allowed to sit
side by side with the king; "where the two good ones sat, uncle
and nephew : as yet was there peace between them, and each
was true to the other4." Hrothgar's sons on the other hand
have to sit with the juniors, the giogoth6. About Heoroweard,
the son of the elder brother Heorogar who had reigned before
Hrothgar, little is said expressly in Beowulf: but it is evident
that he too has been overshadowed.
Everywhere we see, hanging in the background, like a black
thunder cloud, the rivalry which is to arise between the too-
1 See Bugge, H.D. 153-8, 181; Boer in P.B.B. xxn, 377-8; Sarrazin in
Engl. Stud, xxm, 233-5 ; xxvm, 411 ; Chadwick, 300. The conjecture that the
Hothbrodd of Saxo and of the Helgi lays is a personification of the Heatho
beardan was made independently by Bugge, Boer and Sarrazin. It seems
exceedingly probable, in view of the way in which kingly and tribal names
interchange. Of. Ostrogotha, Hoeing, Hunding, Scylding.
2 Enumerated in order of age, as is proved by 11. 465-9.
3 Literally "for the longest time," lengest. See note to 1. 45.
* 1163-5. 5 1188-91.
Stories known to Widsith : Tales of the Sea-folk 83
mighty Hrothulf and his young cousins. Equally with the poet
of Widsith, the poet of Beowulf cannot mention Hrothulf and
Hrothgar together without foreboding evil :
Hrothgar and Hrothulf. Heorot was filled full of friends : at that time
the mighty Scylding folk in no wise worked treachery1.
But the impending evils are alluded to most clearly in the
" tragic irony " with which the Beowulf poet makes Hrothgar 's
queen refer to Hrothulf and his future bearing toward her sons.
Addressing Hrothgar, she says :
I know that my gracious Hrothulf will support the young princes
in honour, if thou, friend of the Scyldings, shouldst leave the world sooner
than he. I ween that he will well requite our children, if he remembers
all which we two have done for his pleasure and honour, being yet a
child.
Then turned she to the bench where her children were, Hrethric
and Hrothmund2.
The story has been lost: but fitting together one obscure
allusion in Saxo, which he himself reproduced without under
standing3, with a name in an old genealogy, we find, almost
beyond doubt, that Hrothulf not only deposed but also slew this
young Hrethric.
It is in contrast with this tragic background that the poet
of Beowulf emphasizes the generous confidence towards each
other of Beowulf and his kinsman, king Hygelac. After
Beowulf, like a good thane, has presented to his lord the
spoils he has just won, the poet moralizes, "So should a
kinsman do : and not weave a net of malice for another with
secret craft : nor prepare death for his comrade in arms." Such
comment seems harsh, and the allusion to treachery uncalled
for, until we notice what that present is which Beowulf has just
given to his lord. It is a war-panoply, which of old belonged
to Hrothgar's brother, King Heorogar, but which has not been
given to Heoroweard, Heorogar's son. No : the armour has
been given to Beowulf the stranger, and Heoroweard has been
deprived of his father's weapons. Small wonder if later he is
1 1017-19. 3 1180-9.
3 Saxo, n (ed. Holder, p. 62). Eoluc is referred to as Qui natum Bjki Eiricum
stravit avari : the reign of Boricus is placed by Saxo after that of Boluo (Books
in and iv) : Saxo does not connect him with Befricus son of B#kus. In Lang*
feftgatal we have Hrarekr Hnauggvanbaugi son of Ingiald, and Hrcerekr
Slaungvanbaugi son of Halfdan. See Langebek i, 5.
6—2
84 Widsith
contemptuously thrust aside by Hrothulf. But Heoroweard
had his revenge. Saxo and the Saga of Rolf Kraki tell us
how Hiarwarus (HjgrvarSr) treacherously plotted against Roluo
(Hrolfr). Saxo did not know who Hiarwarus was, or what was
the ground of his quarrel with Roluo: but he tells how, by
night, after a banquet, he and his men rose upon Roluo and his
warriors, and how Roluo's loyal men fell around him, fighting in
the burning hall.
We cannot rightly understand Beowulf unless we realise the
background against which the hero is depicted. The poet
meant Beowulf to stand out in contrast to the masters of
Heorot, a house of heroes second to none in all northern
story, but tainted by incest and the murder of kin almost
beyond the measure of the lords of Thebes or the house of
Pelops. So he depicts in his hero loyalty, duty, subordination
of his own fortunes to those of his chief.
Widsith shows us that it was this terrible and tragic tale
which the old poets associated with the name of the hall Heorot,
rather than that figure of Beowulf the monster-queller which
the later poet has chosen to paint in the foreground1.
Off a of Angel.
Offa weold Ongle.
The essential part of the tale of Offa is the story of his duel,
while yet a boy, at Fifeldor : the river Eider2. In Beowulf we
find another story connected with Offa's name : that of his
1 See especially a short but most valuable note by Sarrazin in Englische
Studien, xxiv, 144-6 : Rolf Krake und sein vetter im Beowulfsliede ; and Axel
Olrik, Danmarks Heltedigtning, 1903, pp. 11-27. Amidst so much that is con
jecture, the identification of Healfdene, Hrothgar, Halga, and Hrothulf with
Halfdan, Hroar, Helgi, and Hrolf may be accepted as certain. Three out of
four of the names correspond exactly, and the fourth approximately: the
relationships correspond, and half-a-dozen subsidiary marks of identification
are forthcoming. Conjectures, however ingenious, which overlook this are
simply in the air (e.g. the article on Hrothulf by Wilbur C. Abbott in M.L.N.
xix, 122-5, 1905, cf. conclusive answer by Klaeber, M.L.N. xx, 9, 1905).
It seems almost equally incontestable that the Rtfricus and Hiarwarus of
Saxo are the Hrethric and Heoroweard of Beowulf, and that the Beowulf poet
foreshadows the quarrel between Hrothulf and his kin. To my mind the
assertion made twice in Beowulf and once in Widsith that 'as yet' or 'for
a very long time ' or ' at that time ' there was peace within the family implies
necessarily that at last the peace was broken. For a contrary view see Clark,
100.
2 See note, 1. 43.
Stories known to Widsiih : Tales of the Sea-folk 85
queen Thryth1, and her fierceness which none but he could
tame. Whether this story belonged from the first to Offa of
Angel we cannot say ; later, it became attached to the name of
Offa the Second, the historic tyrant-king of Mercia, who, if
pedigrees are to be trusted, was descended in the twelfth
generation from the legendary Offa of Angel. But the in
vestigation of the Thryth story may be left to the student
of Beowulf*: we have no evidence that it was known to
Widsith.
In Beowulf Offa is mentioned as famous in war, as one who
held his land with wisdom ; but the story of his fight at
Fifeldor is not directly referred to.
It is another sign of the close connection between English
and Danish saga that for the details of this story we must turn
to two Danish historians of the end of the twelfth century,
Sweyn Aageson and Saxo Grammaticus. The following is a
summary of Saxo's story, which is the fuller and better of the
two.
Wermundus, the son of Vigletus, king of Denmark, after a long and
prosperous reign begat in his old age his only son Uffo, who grew up tall
beyond the measure of his age, but dull, stupid, and speechless, so that all
men despised him. His father wedded him to the daughter of Frowinus,
the illustrious governor of Schleswig, for from Frowinus and his sons Keto
and Wigo he hoped that Uffo would receive help in ruling the kingdom
after his death. But Frowinus was slain by Athislus, king of Sweden, and
Keto and Wigo brought disgrace upon themselves and their country by
avenging their father in a combat of two against one. Meantime Wer
mundus had grown old and blind, and the king of Saxony laid claim to
Denmark on the ground that its king was no longer fit to rule his realm :
if he would not abdicate, let the quarrel be decided by their sons in single
combat. Wermundus, sighing deeply, answered the Saxon envoys that it
was not just to reproach him with his age, for he had not shirked battle in
his youth : and that, as to his blindness, that deserved pity rather than
insult. He would accept the challenge himself. The ambassadors replied
that their king would not consent to the disgraceful mockery of fighting
with a blind man, and that it would be better that the matter should be
settled by the sons of each. The Danes knew not what to answer : but
Uffo, who by chance was present, asked leave to speak, and " suddenly
from dumb became capable of speech." Wermundus asked who it was
who had thus craved permission to speak, and was told by his retainers
that it was his sou. He replied that it was enough to be insulted by
foreigners, without the taunts of his own household. But when they
1 For the form of the name see J. M. Hart in M.L.N. xvni, 118 ; but cf.
Klaeber in Anglia, xxvm, 448-452 ; Holthausen in Z.f.d.Ph. xxxvn, 118.
2 A detailed study of both stories by Miss E. Bickert will be found in Mod,
Philol. H, 29-76, 321-376.
86 Widsith
insisted that it was Uffo, " Let him speak," he said, " whoever he be."
Then Uffo said, "In vain does the king of Saxony covet the land of
Denmark, which trusts to its true king and its brave nobles : neither is
a son wanting to the king nor a successor to the kingdom." So saying he
offered to fight not only the Saxon prince but, at the same time, any
champion the prince might choose from his nation : hoping in this way to
wipe out the disgrace left upon his country by the murder of Athislus,
perpetrated by Keto and Wigo.
The Saxon envoys jeered, accepted the offer, and withdrew. The
Danes were astonished, and Wermundus praised the valour of the speaker.
He could not believe him to be really Uffo, but he said he would rather
S'eld up the kingdom to him, whoever he might be, than to the proud foe.
ut, when all assured him that it was his son, he asked him to draw near,
and passing his hands over him, recognised his son by the size of his limbs
and by his features....
But when arms were brought, no coat of mail could be found large
enough to fit Uffo, who split the rings by the breadth of his chest : even
his father's was too small. Wermundus accordingly ordered it to be cleft
down the left side and fastened with a clasp, thinking it of small moment
if the side protected by the shield lay open to the sword.
Several swords were offered to Uffo, but there was none so well
tempered that he did not shatter into fragments upon first wielding it.
Now the king had a sword of exceeding sharpness, called Skrep, which
could cut, at one blow, through any obstacle whatever. This sword he
had buried deep, despairing of his son's improvement, and wishing to deny
the use of the sword to others. When he was now asked whether he had
a sword worthy of Uffo's strength, he replied that he had, if he could
recognise the appearance of the place, and find the sword of old committed
by him to the earth. He then commanded his attendants to lead him out
into the open country, and, enquiring the landmarks from them, found the
place, drew forth his sword, and gave it to his son. The sword was frail
from age, and Uffo refrained from testing it; for Wermundus told him
that if he broke it there was no other left strong enough for him.
So they sought the place of combat, an island in the river Eider1;
Uffo alone ; the son of the Saxon king accompanied by a famous champion.
Crowds lined either bank; Wermundus stationed himself at the end of a
bridge, meaning to throw himself into the river should his son be slain.
Uffo parried the blows of his assailants with his shield, but from distrust
of his sword he hesitated to strike, till he discovered from which of the
two he had most to fear, so that him at least he might reach with the one
blow. Wermundus, attributing this hesitation to helplessness, drew
himself gradually to the edge of the bridge. Meantime Uffo first urged
the prince not to let himself be surpassed by his low-born comrade, and
then, to test the courage of the champion, called upon him not to keep in
the rear of his lord, but to repay the confidence which had been placed in
him. Then, when the champion was driven by shame to come to closer
quarters, Uffo clove him asunder with one stroke. Restored by the sound
of the blow, Wermundus cried that he had heard his son's sword, and
asked where it had fallen. And when his attendants told him that the
blow had not pierced any one part but the man's whole structure, he drew
back from the edge of the bridge, desiring now life as keenly as he had
1 This place is referred to again by Saxo in a later book (xn ; ed. Holder, 402)
ubi Wermundi filium Vffonem cum duobus Saxonice gentis lectissimis manum
duelli nomine conseruisse proditum est. The site can be identified with that of
the modern Eendsburg (Miillenhoff, Beovulf, 79).
Stories known to Widsith : Tales of the Sea-folk 87
before longed for death. Uffo then urged the prince to avenge his retainer,
and, having brought him to close quarters by these appeals, he smote him
through with the back of his sword, fearing that the edge was too weak
for his strength. Then Wermundus exclaimed that he had heard a second
time the sound of his sword Skrep, and, when the judges announced that
his son had slain both his foes, burst into tears of joy.
So Uffo ruled over both Saxony and Denmark : by many he is called
Olaf and surnamed ' the Gentle ' on account of the modesty of his spirit.
His further deeds have been lost, though, after such a beginning, we may
well believe them to have been glorious1.
Such is the outline of the story which Saxo tells at con
siderable length in his elaborate and artificial Latin. Very
similar in its main outlines, though not so well told, is the
version given in the scanty chronicle of Saxo's predecessor,
Sweyn Aageson2. In Sweyn, while we miss much that is most
picturesque in Saxo's tale, we get a few additional details of
value. Thus Saxo does not seem to have made up his mind as
to the reason of Uffo's dumbness and incapacity. Saxo's Uffo,
after his reply to the envoys, is questioned by Wermundus, "why
he had taken care to cover his sweet eloquence by the most
studied dissimulation and had borne to pass so great a period of
his age without speech," and merely makes answer that he
had been content to live under the defence of his father, and
had had no use for his voice until he had observed the wisdom
of his own land hard pressed by the quick speech of foreigners.
To which Wermundus replies that his son has made a just
judgement in all matters. Yet Uffo's explanation seems
somewhat inadequate. It is Sweyn Aageson who tells us that
Uffo had been rendered dumb by the shame brought upon his
country by two men having combined to avenge their father
upon a single foe : obviously an allusion to the matter of Keto
and Wigo, concerning which we hear at length in Saxo. This
detail, supplied by Sweyn, explains in its turn a detail given
1 Saxo, iv, xxxiv-vi (ed. Holder, 106-115).
2 Langebek, i, 45-47. The story is also given, very briefly, in the Annales
Ryenses :
" Wermundus Blinde, Hie vir bellicosus erat. Hujus tempore Keto, &
Wiggo, filii Frowini Praefecti Sleswicensis, occiderunt Athislum Kegem Svecise,
in ultionem patris sui. Huic successit Uffo Starke filius ejus.
Uffo Starke. Iste a septimo aetatis anno nsqve ad trigesimum noluit loqvi,
qvousqve in loco, qvi adhuc Kunengikamp dicitur, super Eydoram cam filio
Eegis Teutonicorum & meliore pugile totius Teutonise solus certans, ambos
occidit & sic conditione utriusqve gentis, Teutonia Danis jam qvarto tributaria
facta est." Annales Ryenses. In Langebek, i, 152, under the name of Chronicon
Erici Regis.
88 Widsith
only by Saxo : that Uffo's reason for demanding two opponents
is that he may wipe off the disgrace left upon his country by the
shameful act of Keto and Wigo ; a disgrace of which, had we
Saxo's story only, we should have supposed that he had hitherto
taken no heed. It is evident that in the poem which formed
the common source of both accounts this deed of shame was
intimately connected with Uffo's victory, the significance of
which lay, not, as it apparently does in Widsith, merely in the
saving of the land from foreign aggression and the extending
of its borders, but rather in the wiping off, by a deed of
prowess, of the shame left by a dastardly murder.
In Sweyn, Wermundus, when he has felt over the mighty
limbs of Uffo to satisfy himself that it is of a truth his son,
exclaims " Such I remember was I myself in the flower of my
youth ! " Finally in Sweyn Uffo enters the lists girt with his
father's sword, but holding another in his hand. When the
worthless sword breaks asunder the German foes exult, and the
Danes quake with fear; but Uffo suddenly draws his second
sword and fells his adversary. From this it has been argued
that the immediate source of Sweyn was not identical with
that of Saxo ; for Saxo, it is said, and with reason, would not
have knowingly slurred over a situation like this1.
It is, then, from a comparison of the historians that we must
reconstruct the Danish lay. We see that, as in the old English
battle poems, the speeches of the combatants played an impor
tant part. Uffo's words to his two foes, the prince and the
champion, are recorded at length in both accounts.
About the same time that these Danish ecclesiastics were
compiling their histories of Denmark out of old traditions and
war poems, an unknown monk of St Alban's2, in the hope of
glorifying his monastery by commemorating its founder Offa II,
and his namesake and ancestor Offa I, was committing to
writing the traditions of the two Offas, as current in his day3.
1 Olrik, Sakses Oldhistorie, n, 182.
2 Miss Eickert has shown that there is every possibility of the author having
been abbot John de Cella (1195-1214). See Mod. Philol. n, 29 (1904). The
author was not, as is often asserted, Matthew Paris.
8 Solent autem de isto Offa multa narrari...ut dicunt, etc. in MS Cotton
Stories knotim to Widsitli: Tales of the Sea-folk 89
The English story is substantially the same as the Danish :
it is however located in England : all recollection of the
continental Angel has vanished. Warmundus is a king of the
Western Angles, ruling at Warwick. Offa, his only son, is
blind till his seventh, dumb till his thirtieth year. A certain
Riganus claims to be recognised heir in Offa's place, and, when
Warmundus refuses his consent, raises an army. In answer to
prayer Offa receives the gift of speech, harangues the council,
leads his army against the foe, joins battle, dashes with his
men across the river, and in the midst of the combat slays the
two sons of the usurper, fighting one against two. Then
follows the romantic story of Offa's queen : not the tale referred
to in Beowulf, which in the Vitae Duorum Offarum is transferred
to Offa II, but a version of the widely spread motive of the
fugitive princess who is wedded to a great king, of the inter
cepted letters, of the consequent attempted slaughter of the
queen and her children, and of the final reconciliation — in fact
the Man of Law's Tale1. Offa is instructed by a hermit to
build a monastery as a thank offering for his recovery of wife
and children, but he omits to perform his vow, as do all his
house after him till the days of Offa II, to whose life the
monkish biographer then passes. The history of Offa II is a
combination of historic fact with details repeated from the
legend of Offa I, and does not concern us here.
The relationship of the Vita Offae I to the Danish histories
was a subject often discussed during the 18th and early 19th
centuries. Early students necessarily ignored the reference in
Widsith, which was not published till 1826 ; and therefore,
though they are sometimes useful in points of detail, their
conclusions are valueless2. We now know that, five or six
Julius D vn, supposed by Miss Eickert to be an earlier draft by the compiler of
the Vitae ; omnium fere conprouincialium assercio in the Vita Offae I,
These Vitae were published with Wats' edition of Matthew Paris, London,
1640. They have since only been reprinted in extracts.
1 See the Originals and Analogues of Chaucer's Canterbury Tales (Chaucer
Soc.) pp. 1-84 (where part of the Vita Offae I is reproduced), 221-250, 365-414.
2 Bibliographies of the Offa legend will be found, by Suchier, in P.B.B. iv,
500, and by Miss Bickert in Mod. Philol. n, 29. P. E. Miiller's discussion of
the subject in his edition of Saxo (Copenhagen, 1858, Part n, Not. Vb. 137), as
it was made without reference to Widsith, must be ranked with the earlier
criticisms. An interesting Celtic parallel has been pointed out by G. H. Gerould :
Offa andLabhraidh Maen, in M.L.N. xvn, 201 etc. (1902).
90 Widsith
centuries before either Saxo or the monk of St Albans was
writing, the story of the duel on the island of the Eider was
being told of Otfa, king of Angel. The allusions in Widsith
are not detailed enough to enable us to reconstruct fully this
most ancient version : but we can see that it was nearer to the
story, as later preserved in Denmark, than to that preserved by
the monk of St Albans. The quarrel was with an alien and
not with a domestic foe ; the fight took place by the Eider ; it
was a duel, and not a pitched battle ; Widsith shows us that in
all these points the Danish version is the original one. It is
natural that this Danish account should make Offa into a
Danish prince : the fact that he defended his frontiers on the
river Eider necessitated this, at a date when the Eider had
come to be the frontier between the Danes and the German
races of the south. Naturally, too, the foe has come to be
identified in Danish story with these Germans: Saxo calls them
Saxons, Sweyn Alamanni. From Widsith we learn that the
contending parties were really the Angles and Myrgings, who
seem to have been neighbours of the Angles on the south.
There can be little doubt that so much of the story as is
common to English and Danish tradition goes back to an early^
Anglian lay, brought over to England by the Angles, but
surviving also in the neighbourhood of the ancient Angel, and
incorporated by the Danes into their national stock of stories,
just as they incorporated into their kingdom the ancient Angel
and the remnants of its people.
That the English life of Offa and the Danish story of Uffo
were not derived the one from the other, but were independent
traditions from the same original, had been perceived by one
historian1 even before the discovery of Widsith, which has
rendered it certain. More recently Mlillenhon02, it is true, has
argued that the story had no continuous life in Denmark, but
was reintroduced there from England; but his arguments, on
examination, have fallen to the ground3. All the evidence
1 Dahlmann, Forschungen, r, 234, 235.
2 Beovulf, 80 etc.
3 See the criticisms of Miillenhoff's view by Chadwick (125-6) and by Olrik
in the Ark. f. word. Filologi vin (N. F. iv) 368-75 : Er Uft'esagnet invandret fra
England ? Bemeerkninger til Miillenhoff's ' Beovulf.'
Stories known to Widsiih : Tales of the Sea-folk 91
indicates that Saxo's story goes back to a Danish lay, which, in
its turn, was founded upon a tribal lay of the continental
Angel. Many of Saxo's proper names such as Uffo1, Frowinus,
Skrep2, and, above all, the localization of the fight on the
Eider, point to the story having had its original home not in
England but in the old home of the Angles in Schleswig. It
is indeed impossible that Saxo could have known of the
localization on the Eider, if, as Miillenhoff supposed, he drew
from a tradition which had travelled to England in the sixth
century and back to Denmark perhaps in the tenth. On the
contrary, the tradition Saxo used must have handed down the
fame of Offa of Angel continuously in the near neighbourhood
of those lands where Offa had fought and ruled.
Yet we must not suppose that the tale as known to Widsiih
was in all respects that which Saxo has preserved for us. We
cannot exactly gauge what has been the loss and the gain of
the centuries intervening between Widsith and Saxo. We
may safely suppose, though the scanty allusions of Widsith do
not tell us so, that the sluggish youth of the hero was an
original feature of the story. The fact that this trait is
preserved in the twelfth century English and Danish versions
alike, proves that it must have occurred in the common
original of both ; and we know from Beowulf3 that the sluggish
hero who proves in the end a good champion was as well known
a character in Old English as he was in Middle English story.
It is puzzling, however, to find that, in Widsith, Offa wins his
honours while still a youth (cniht wesende) whilst the later
versions, both English and Danish, agree in making the period
of torpor last till his thirtieth year4.
1 Chadwick (126-7).
2 Koegel, Ltg. , i, i, 162-3. " The name of the sword skrep is not Danish,
and not Norse as Uhland thought (Schriften, vn, 215) but Anglian. If any
further proof were needed that the ultimate source of Saxo's story is an Anglian
poem, composed by the countrymen of Offa, and dating at the latest from the
beginning of the sixth century, it would be afforded by this word. For skrep
is the same as the West Frisian schrep, the Low German schrap, firm, steadfast
(Doornkaat Koolmann, 3. 145 a). The name bears upon the story, for Skrep is
the only sword which is sufficiently firm and steadfast to endure Offa's test."
3 1. 2187. For Miss Eickert's interesting suggestion that we should retain
the MS reading geomor in Beowulf as a reference to Offa's heaviness, see Mod.
i, 54-58. « See note to 1. 39.
92 Widsith
What gives its dignity to the Danish story is the connection,
which Saxo himself only partly understood, between Uffo's
combat and the deed of shame wrought by Keto and Wigo.
By this connection the interest of the story centres no longer
in any single person, but in the honour of Uffo's people1. One
would like to think that this was a feature of the original
Anglian story. We cannot be sure that it was: yet it may well
have been, for Freawine and his son Wig were known to
English tradition. We find them in the West Saxon pedigree
among the ancestors of Cerdic2. And we shall find that Widsith
seems to show some trace of a knowledge of their story.
Perhaps, finally, we should not be wrong in attributing to
the ultimate Anglian source that quality of straightforward
narrative in virtue of which Saxo's story of Uffo stands out
from most of his history. Koegel3 has claimed for this episode
an ' epic breadth,' such as we find in Old English, but not in
Old Scandinavian heroic verse. Every reader of Saxo must
feel the difference between the manner in which this story is
told and the lyrical treatment which is given, for example, to
the story of the death of Rolf Kraki, where we know that Saxo
was following a Scandinavian lay. But it may be dangerous to
argue too much from this difference of treatment.
Eadgils.
The story of the slaying of Athislus, king of Sweden, as
told by Saxo, is briefly this. He invaded Schleswig, defeated
Frowinus, the governor, and slew him in single combat. Upon
his making a second attack, Wermundus, assisted by Keto and
Wigo, the sons of Frowinus, drove him back: but Athislus
escaped with his life, although he bore the brunt both of the
attack and of the retreat. Keto and Wigo journeyed to
Sweden, and, finding Athislus alone, challenged him to combat;
but he, pitying their youth, offered to pay the money com
pensation for their father's death. Keto refused the offer, and
1 Olrik, Sakses Oldhistorie, n, 185.
2 Olrik thinks the episodes were originally distinct, ' ' Oprindelig har de
to begivenheder nsappe htfrt sammen " (184); but the reasons he gives are not
conclusive. For Freawine see Grimm, trans. Stallybrass, i, 211.
3 Ltg. i, 1, 159.
Stories iknmvn to Widsith : Tales of the Sea-folk 93
attacked Athislus, bidding his brother stand aside, "for the men
of old time held it unjust and scandalous for two men to fight
against one, and counted such a victory as a disgrace." Athislus
for some time held back, bidding the two youths come on
together ; till the vigour of Keto's attack compelled him to put
forth his full strength, and he struck him to his knees. Then
Wigo " made affection conquer shame " and interfered to save
his brother: and together they slew Athislus.
Now we have much information, half legendary, half historic1,
about the kings of Sweden, dating from the days, three centuries
earlier than Saxo, when Thiotholf the scald made his Ynglinga-
tal for king Rognvald Higher-than-the-Hills. But we know
of one Swedish king alone named Athils, the son of Ottar (the
Eadgils, son of Ohthere, of Beowulf), who gained the king
dom by the rout of Ali (Onela)2 and whose dealings with Rolf
Kraki (Hrothulf) were such a favourite subject of Scandina
vian story3.
But this Athislus-Eadgils cannot be the Athislus who was
slain by Keto and Wigo : for our authorities are agreed that he
died, not in fight, but whilst celebrating a ceremony :
King Athils was at a sacrifice of the goddesses, and rode his horse
through the hall of the goddesses : the horse tripped under him, and fell,
and threw the king ; and his head smote a stone so that the skull broke,
and the brains lay on the stones, and that was his death. He died at
Uppsala, and there was laid in mound, and the Swedes called him a
mighty king4.
Since there is no other Swedish king of the name, Saxo
must be wrong in supposing the king Athislus who was slain
by Wigo to have been a Swede. But in that case he is most
probably identical with Eadgils, lord of the Myrgings, in
Widsith. For the Myrgings, we gather from Widsith, were
probably also called Swcefe5, and, in view of the frequent
1 Chronologically the Ynglingatal is remarkably consistent. Cf. Heusler,
Zeitrechnung im Beowulfepos in Herrig's Archiv, cxxiv, 9.
2 Beowulf 2392 etc. Ynglingasaga in Heimskringla, i, 29. Sn. Edda, Skald.
41 (44), udg. Jdnsson p. 108.
3 Hrolfs Kraka saga, especially cap. 28 etc. ; Heimskringla, Sn. Edda, as
above ; Saxo n (ed. Holder, pp. 54-5).
4 Heimskringla i, 29. Saxo has a different account. Athislus brought on
his own death through celebrating, with immoderate drinking, the funeral of
his enemy Eoluo (Book in ; ed. Holder, p. 75).
5 See note to 1. 44.
94 Widsith
mediaeval confusion of Sweden and Swabia, a king of the
Swaefe is most likely of all men to have been made into a king
of Sweden. Besides, no other Athils-Eadgils is, so far as
I know, forthcoming to contest this place with Eadgils the
Myrging1. But if Old English tradition represented the Eadgils
slain by Wig as a Myrging, and Offa as defending his country
against an attack of the Myrgings, it becomes highly probable
that the two stories were even more closely connected than we
find them in Saxo. Eadgils the Myrging slays Freawine, an
Anglian chief: Ket and Wig avenge their father: Offa repels
the avenging Myrgings and restores the honour of the Angles
by his duel at Fifeldor.
This story is probably based upon events which actually
happened in the later half of the fourth century of our era2.
Much further back than this we can hardly trace English
history. Wermund's father is Wihtla3g in the O.E. genealogies,
Vigletus in Saxo : these may be identical, and point to some
hero of history or legend. Beyond WihtlaBg we have only
Woden in the English genealogy. Saxo makes Vigletus slay
Hamlet, and wed Hamlet's wife ; but we have no evidence that
this Jutish saga was known in England.
1 That the Athislus of Saxo (Book iv) is identical with the Eadgils of
Widsith seems of recent years to have occurred independently to at least three
scholars : Sarrazin in Engl. Stud, xxiu, 234 ; Chadwick (135) ; Weyhe in Engl.
Stud, xxxix, 37. Sarrazin and Chadwick both point out how easy and how
frequent is the confusion of Swedes and Swabians. But the identification of
the Athislus of Saxo with the Eadgils of Widsith, and of the Swedes of Saxo with
the Swcefe of Widsith, was first proposed at least sixty years ago (by P. A.
Munch, Det norske Folks Historic, 1851, i, 1, 244).
2 For this date Chadwick (p. 136) relies largely upon Widsith, which represents
Eadgils as a contemporary of Ermanaric (died c. 375). But Widsith equally
represents him as a contemporary of Alboin (died c. 573) and on this ground
Eadgils used to be placed with equal confidence in the sixth century. In fact,
no chronological arguments can be safely drawn from Widsith. However, as
Chadwick points out, if we reckon the generations back from the earliest figures
which we can date in the Mercian and West Saxon pedigrees, allowing thirty
years for a generation, we should have to put the birth of Wig c. 350, and that
of Offa c. 360. This would agree excellently with the story in Saxo. The West
Saxon pedigree is not reliable, but the Mercian probably is, and similar results
have been drawn from it by Mullenhoff, (Beovulf, 85-6.)
Stories known to Widsiih : Tales of the Sea-folk 95
The tale of Wade.
Wada [weold] Hcelsingum.
Perhaps Wade was originally a sea-giant, dreaded and
honoured by the coast tribes of the North Sea and the Baltic1.
That he retained something of this character in these regions
till the thirteenth century is proved by the references to him
in the Thidreks saga, where the giant Vathe is the offspring of
King Villcinus and of a sea-wife. The saga has, however,
little to tell concerning him ; he owes his place in the story to
his being the father of the more interesting Weland : a
connection which is no doubt late. On one occasion " a kind
of heathen Christopher " he wades over the swollen sound with
his son upon his shoulder2. Vathe is finally overwhelmed in
an avalanche3. Perhaps we may see in this a survival of some
legend of the storm-divinity.
We meet with Wade again in a High-German disguise in
the poem of Kudrun, almost contemporary with the Thidreks
saga. Here he has been brought into subjection to the laws
of chivalry, and becomes the type of the faithful retainer, but
his old aquatic habits peep forth. To him the ways of the sea4
and the tales of the sea5 are known : and when the angel tells
Kudrun that her deliverers are at hand, it is Wade who steers
the fleet which brings them :
dir kumet in ditze lant
Wate von den Stiirmen ; der hat an slner hant
ein starkez stiurruoder6.
In England the memory of Wade lived longer than that of
any of the old heroes of song, Weland only excepted. The
references to him by the two greatest writers of the Middle
English period have kept his name from being ever quite
1 Miillenhoff in Z.f. d. A. vi, 62; Symons in Pauls Grd.(3) in, 718-9. It
is not necessary, with Miillenhoff, to assume a Frisian or Prankish origin for
Wade ; evidence would rather lead to our placing both him and his people on
the Baltic, very near the old Anglian home. See note to Hcelsingas, 1. 22, and
cf. Golther, 176.
2 cap. 58. 3 cap. 60.
« Stanza 836. 8 1128. • 1183.
96 Widsith
forgotten. In Chaucer's Troilus a " tale of Wade " helps to
while away an evening, when Pandare entertains Criseyde1 :
And after semper gonnen they to ryse
At ese wel, with hertes fresshe and glade,
And wel was him that coulde best devyse
To lyken hir, or that hir laughen made.
He song ; she pleyde ; he tolde tale of Wade.
Elsewhere Chaucer refers to Wade's boat2.
In Mallory's Morte d' Arthur the damsel Li net says to Sir
Gareth of Orkney:
For were thou as wight as ever was Wade, or Launcelot, Tristram, or
the good knight Sir Lamorake, thou shalt not pass a pass here, that is
called the pass perilous3.
References in less famous works are fairly frequent. Sir
Bevis of Hamton classes Wade amongst the few champions of
romance who have slain their dragons4. We shall probably be
right if we regard this dragon episode5 as a later development
of his story.
In the alliterative Morte Arthure the " wery wafulle wedowe "
warns King Arthur that it is useless for him to attack the
giant :
Ware thow wyghttere thane Wade or Wawayne owthire,
Thow wynnys no wyrchipe, I warne the before6.
Even in 1598, when Speght edited Chaucer, the name of
Wade's boat was remembered:
Concerning Wade and his bote called Guingelot, as also his strange
exploits in the same, because the matter is long and fabulous, I passe it
over7.
Speght has been denounced by Tyrwhitt8, and by Sir
1 Bk m, 1. 614, ed. Skeat. For further notes on Wade, see Grimm, Kl.
Schr. vni, 502, Kemble, Saxons, i, 419-20. The suggestion made by Kemble in
his footnote is quite untenable.
2 Marchantes Tale, 180 (T. 9298).
3 Book vii, cap. EC.
4 Sire Launcelet de Lake,
He faugt wi\> a fur drake,
And Wade dede also.
Sir Beues of Hamtoun, ed. Kolbing, E.E.T.S. 2603-5.
B Of. Binz in P.B.B. xx, 197 ; Panzer H.G. 289.
6 Ed. Brock, E.E.T.S. 1865. Another reference to the romance of Wade
will be found quoted in Warton's History of English Poetry, 1774, vol. i, p. 120.
7 The Workes of Chaucer, 1598. Annotations.
8 Tyrwhitt's Canterbury Tales, W. Pickering, 1830, iv, 248.
Stories known to Widsith : Tales of the Sea-folk 97
Walter Scott, for having " to the great prejudice of posterity1 "
given us only this tantalizing reference. But probably he knew
little of the story. He wrote more than a century after the
latest of the references given above, and it is unlikely that,
except locally, any tale of Wade was remembered, in the
year 1600. To the antiquary Leland, much earlier, the name
Wade had no meaning. Leland notes that at Mulgrave, near
Whitby, are " certen stones communely caullid Waddes Grave,
whom the people there say to have bene a Gigant2." Camden
also, a contemporary of Speght, visited Wade's grave, but the
name does not seem to have awakened in him any recollection
of old tradition. He connects the eleven foot high giant with
the historic Northumbrian chief named Wade3. Finally, Sir
Francis Kynaston, in his commentary upon Troilus(c. 1635), pro
fessed to give an account of Wade, which shows that he could
find nothing to add to what Speght had condescended to state :
Tale of Wade, &c. Chaucer means a ridiculous romance, as if he had
told a story of Kobin Hood, for in his time there was a foolish fabulous
Legend of one Wade & his boate Guingelot, wherein he did many strange
things & had many wonderfull adventures ; not much unlike that man and
his boate in our time, who layed a wager, that he never going out of his
boate, & without any other helpe but himselfe, he would in a certaine
number of dayes go by land & by water from Abington to London, & in
his passage would go over the top of a square Steepel by the way, which
thing he performed & wonne his wager4.
It is evident, then, that by the early seventeenth century
the tale of Wade was forgotten.
Locally, Wade's name was remembered much later. At
Mulgrave the grave of the giant Wade was shown, almost down
to our days. Also, says an antiquary of the late eighteenth
century :
The common people... shew... a huge bone which they affirm to be
one of the ribs of Bell Wade's cow ; believing her, as well as her owner, to
1 Lay of the Last Minstrel, note 2 E. Compare Scott's introduction to Sir
Tristrem, 1804, p. Ixi (1806, p. Iviii).
2 Leland's Itinerary, second edit., Hearne, 1745, i, 60.
3 Hie in colle inter duo saxa solida septem plus minus pedes alta tumulatur,
quae cum vndecem pedibus distarent ilium gigantea fuisse mole non dubitant
affirmare. Britannia, Londini, 1594, p. 555. See Gough's Camden for further
notices of Wade, especially " Wades gap."
4 This commentary has never been published. The extract above is taken
from The Loves of Troilus and Creseid, written by Chaucer ; with a commentary,
by Sir Francis Kinaston: never before published [edited by F. G. Waldron] 1796.
Only the first part was issued.
c. " 7
98 Widsiih
have been of an enormous size. But to me this appeal's a mere fiction ;
for the bone now preserved there has doubtless belonged to some fish,
probably a whale ; though such a suggestion is by no means agreeable to
its present possessors1.
The name of " Wade's causey " was also given to a disused
Roman road
which leads from it [the village of Dunsley, on the sea coast] many miles
over these vast moors and morasses toward York. This extraordinary road,
at present disused, is called by the country-people Wade's causey, concerning
which they relate a ridiculous traditional story of Wade's wife, and her
cow. The fabulous story is, that Wade had a cow, which his wife was
obliged to milk at a great distance, on these moors ; for her better con
venience he made this causeway, and she helped him by bringing great
quantities of stones in her apron ; but the strings breaking once with the
weight, as well they might, a huge heap (about twenty cartload) is shown
that dropped from her2.
Such is the last echo, on the Yorkshire coast, of the tales of
the dread sea giant which the conquerors had brought with
them across the North Sea thirteen centuries before.
The name Wade was frequently borne in Old English times
by historic chiefs3. We cannot therefore feel sure that the
many places associated with the name4 all point to a localization
of stories of the mythical hero. It is particularly tempting to
assume this, however, when we find the name associated with
mills and bridges6.
Of the lost Tale of Wade six lines were discovered by Dr
Montague R. James in a Latin sermon on Humility, of the
early thirteenth century. As emended and explained by Prof.
Gollancz, they run :
2ta quod dicere possunt cum Wade:
Summe sende ylues
and summe sende nadderes :
summe sende nikeres
the bi den watere wunien.
Nister man nenne
bute Ildebrand onne6.
The allusion to nickers dwelling by the water seems
1 The History of Whitby, by Lionel Charlton, York, 1779, p. 40.
3 The hittory and antiquities of Scarborough and the vicinity, by Thomas
Hinderwell, second edit., York, 1811, p. 19.
8 About a dozen Wades are recorded in Searle's Onomasticon.
4 For a list of these see Binz in P.B.B. xx, 198.
8 For instances of these see Francesque Michel, Wade, Lettre sur une
tradition angloise du moyen age, Paris et Londres, 1837. For German examples
of Wade in place names, Watanbrunno, etc., see Mullenhoff in Z.f.d.A. vi, 65.
6 See Academy, 1241, Feb. 1896, p. 137; Athenaum, 3565, Feb. 1896, p. 254.
Stories known to Widsiih : Tales of the Sea-folk 99
appropriate to the tale of Wade. But this reading is con
jectural, and the lines still leave us in the dark as to Wade
and his boat Guingelot.
Perhaps more help is to be got from the story of Gado,
told us by Walter Map1.
Qado was the son of the king of the Vandals ; but, from love of
adventure, left his home when a boy. A mighty hunter, versed in all
knowledge, he wandered through the world redressing wrongs, and at last
he came to the court of King Offa, the builder of the Welsh dyke. Offa,
to his sorrow, wedded the daughter of the Koman emperor ; his covetous
Eoman guests, returning home, urged the emperor to make Offa tributary,
and were deterred from attacking him only by fear of his friend Gado.
However, when Gado was called off, to right wrongs in the furthest Indies,
and had been dismissed by Offa with rich gifts, the Romans, with a mighty
army, attacked Offa unawares. But Gado, having righted things in India,
was returning to his native land : when against his wish, his ship was
carried to England, near Colchester. There Offa had come to meet the
Romans, who had just landed, and who had refused his terms of peace.
After greeting Offa, Gado, clothed in his wonted glorious vestments, and
accompanied by a hundred chosen knights (for he always led with him at
least a hundred), went to the Roman quarters to act as mediator, but was
repulsed. So Gado arrayed the English forces, placing Offa with the main
body in the market place of the town, the king's nephew Suanus with a
chosen five hundred at one exposed gate, and himself, with his hundred,
at the next.
Fearing Gado, the Romans concentrated all their attacks upon Suanus,
who, at the third assault, had to appeal to Gado for help. But Gado
refused, and Suanus was preparing to sell his life dearly, when Gado
commanded him to draw back. He obeyed: the enemy rushed in, and
were met by Offa in the market place, whilst their retreat was cut off by
Gado from behind. Thus caught, the Romans were mowed down, till
quarter was offered to the survivors, who returned to Rome with their
Map has rationalized the story, and has toned down, without
obliterating, its supernatural features. The boat which enables
Gado to go to the uttermost Indies, to accomplish his adventure
there, and to return in time to thwart the emperor, a boat which
carries him magically to the spot where there is work for him to
do, can be no other than Guingelot. Gado's wisdom, and his white
hair8 (characteristics which are also preserved in Kudruri), are
probably derived from the original figure of the dreaded old
1 De nugis curialium, ed. Wright, Camden Soe., 1850, n, 17. The parallel
has been pointed out by Miss Rickert in Mod. Philol. n (1905), and by Brandl
in Herrigs Archiv, cxx, 6, and Pauls Grdr.fa) u, 1, 1085.
3 Compare Omnem enim habere sapientiam aiebant, linguat quorumlibet
loquebatur regnorum, et frequenti felicitate successuum totius vitee videbatur
obedientiam obtinere, tanquam optioni suce parerent animantia omnium motabilium,
et haberent intelligentiam, with the thrice repeated formula Wate der vil wise ;
respersum cants quati semicanum with swie grit er do ware.
7—2
100 Widsiih
man of the sea. In other points the coincidence between the
Anglo-Latin and the High German story may be accidental.
The glorious clothing of Gado and his companions astonishes the
emperor and his knights ; we are reminded of the magnificence
of Wate's companions at the Irish court ; when Gado takes over
the command from king Offa and directs all the fighting around
the town gates, we are reminded of the similar commanding
part assumed by Wate at the end of the Kudrun story.
That Gado should be the son of a Vandal king carries us
back to the flourishing days of the Old English heroic story,
when the Vandal name was clearly remembered ; this feature
would hardly have been invented later. It was probably the
exigencies of alliteration that originally caused a Vandal king
to be chosen for Wade's father, just as in the Thidreks saga he
is made the son of king Villcinus.
Putting together, then, the accounts of Wate and Vathe, of
Wade and of Gado, we find these common characteristics, which
we may assume belonged to their ancient prototype, Wada
of the Haelsingas:
(1) Power over the sea.
(2) Extraordinary strength — often typified by superhuman
stature.
(3) The use of these powers to help those whom Wade
favours. Not that he is necessarily gracious or benevolent.
Originally he was perhaps the reverse1. But probably he grew
out of the figure, not of a historic chief, but of a supernatural
power, who had no story all his own, and who interested mortal
men only when he interfered in their concerns. Hence he is
essentially a helper in time of need : and we may be fairly
confident that already in the oldest lays he possessed this
character.
Hagena and Heoden.
Hagena [weold] Holmrygum and Heoden Glommum.
In these words we have a reference to one of the greatest
1 This sinister character he keeps in the Thidreks saga ; cf . Harm var illr
vtiSrskiptis, oc firir j>a scee var hann o\>ockas<ell. Wate, too, has something
of this grim nature.
Stories known to Widsith : Tales of the Sea-folk 101
of the old tales : a tale so popular that allusions to it are to
be found everywhere, and that four complete, although late,
versions have come down to us.
The best of these is the Icelandic version as given by
Snorri1, in explanation of the allusions of the poets:
A battle is called the blast or shower of the Hjathningar [O.E.
Heodeningas], and arms the fire or wands of the Hjathningar, and there
is a tale as to this.
The king named Hogni [O.E. Hagena] had a daughter named Hild ;
the king named Hethin [0. E. Heoden] son of Hjarrandi [0. E. Heor-
renda] carried her off by force. At that time Hogni was gone to a meeting
of chieftains, but when he heard that his kingdom had been raided, and
his daughter carried off, he went with his company in search of Hethin,
and learnt that he had gone North, coasting the land. And when king
Hogni reached Norway he learned that Hethin had sailed West, over
the sea. Then Hogni sailed after him to the Orkneys ; and when he came
to the island called Hoy, Hethin was there, with his company.
Then Hild went to meet her father and offered him a necklace 2 as a
peace offering, on behalf of Hethin. But on the other hand she said that
Hethin was ready to fight, and Hogni need expect no forbearance from
him. Hogni made answer sternly to his daughter : but she, returning to
Hethin, told him that Hogni would have no peace, and bade him prepare
himself for battle. And so they did on both sides ; they went up on
to the island, and drew up their companies.
Then Hethin called to Hogni, his father-in-law, and offered him
peace, and much gold in atonement. But Hogni made answer, " Too late
dost thou proffer peace, seeing that I have now drawn Dainsleif, which the
dwarfs smithied : and it must be the death of a man every time it is
drawn, nor does it ever fail in its stroke, nor may a scratch from it be
healed." Then said Hethin : " There thou dost boast of thy sword, but
not of the victory : I call that good which is true to its lord."
Then they began the battle, which is called Hjathningavig ; and they
fought all the day, and in the evening the kings went to their ships.
But Hild went by night to the corpses, and awoke the dead by magic.
And the next day the kings went to the battlefield and fought, and so did
all those who fell the day before.
In such wise the battle continued, day after day ; so that all those who
fell3, and all the weapons and shields which lay on the battlefield, were
turned into stone. And when it dawned, all the dead men stood up and
fought, and all the weapons were sound : and it is told in songs that the
Hjathningar shall so abide till Doomsday. And of this tale Bragi the
poet wrought in the song of B&gnar Lodbrok.
This is the same song of Bragi which is our oldest Norse
authority for the tale of Ermanaric4. We must not, however,
argue that the Hild story was known to Bragi in the ninth
century exactly as Snorri tells it in the thirteenth. The
references in the Hdttalykill of Rognvald, earl of the Orkneys,
1 Sn. Edda, udg. J6nsson, SkdldsLaparmdl, 47 (50).
9 Beading, with J6nsson, men at scett. But cf. Panzer, H.G. 160.
8 Cf. -Panzer, H.G. 160 (footnote 2). 4 See above, pp. 19-20.
102 Widsith
which are much nearer to Snorri's time than Bragi's references,
show a version of the story not quite consistent with Snorri's,
and in the Sorlathdttr1 the tale is told again with many
additional details, not all of them to be attributed to the
fourteenth century compiler.
But, however they may differ in detail, all these Norse
versions bear a common stamp2. In them all the essential
feature is the tragic fight between father-in-law and son-in-law,
following a vain attempt at reconciliation. How Hethin gained
the love of Hild we are not told8.
But at the same time that Snorri was writing his "poets'
manual " in Iceland, the story was being made the theme of epic
song in Austria. In this southern version the essential part of
the tale is the wooing of Hild : the battle between father- and
son-in-law, which follows, ends in a reconciliation. Two striking
features of the High German epic are the griesly old warrior
Wate [Wade], the irresistible servant and protector of Hetel
[the M. H. G. equivalent of Hethin or Heoden] , and the sweet
song of Hdrant, by means of which the love of Hild is won.
This southern version is represented to us by Kudrun, a
High German poem founded upon Low German tradition. After
the story of the youthful adventures of Hagen — a later develop
ment of the tale — we are told how Hagen rules in Ireland, and
hangs all the messengers sent to beg for his daughter's hand :
conduct which nevertheless does not check the crowd of wooers.
The wise men of King Hetel of the Hegelings counsel him to wed.
Hetel consults his vassal Horant as to his suit for Hild : only by the help
of Wate, Horant says, can the lady be won. Wate is summoned : and it
is decided that he and Horant, with other champions, shall sail in dis
guise to Ireland, and carry off the bride. They are to feign themselves
1 Fornaldar Sggur, ed. Bafn, i, 389-407, §§ 5-8. It is to be noted that this
version makes Hogni and Hethin foster brethren. Hethin is left in charge of
Hogni's land : an enchanted drink causes him to break his trust, murder Hogni's
wife, and carry off his daughter.
2 The versions and allusions were compiled by P. E. Muller in his edition of
Saxo, Notce uberiores (vol. n, pp. 158-61), and more fully by Prof. Piper in
his edition of Kudrun (Anhang zur Einleitung, 60-135). This collection does
not however include the Hdttatykill, an extract from which will be found in
Panzer, H.G. 171.
3 It must be remembered, however, that Snorri tells the story in order to
illustrate a particular phrase. He may have known more of the earlier part of
the story than he thought it worth while to tell.
Stories known to Widsith: Tales of the Sea-folk 103
merchants, or else exiles outlawed by Hetel ; but with armed men con
cealed in the hold. Both disguises are combined (the poet is apparently
harmonizing two dissimilar traditions1) ; the magnificent merchants are
invited to Hagen's court. Wate excites first terror by his grim appearance,
then admiration by his skill in swordsmanship. Equally admired is
Horant's morning song, which shames into silence the birds singing in the
bushes2. Hild is charmed by it ; she cannot rest till Horant comes to sing
to her in her chamber. So Horant sings a song of Amile" : the like of
which was never known by Christian man before or since, unless he heard
it upon the wild waves.
Then he throws off his disguise and woos for his lord. " Noble maid,
my lord has in his court twelve who sing far beyond me : and all so sweet
as is their song, yet my lord sings best of all."
The maiden is won, and consents to flee with the wooers. So Wate
announces to Hagen that the ban has been removed, and he and his
companions have been summoned home. He asks the king to do them
one last favour by visiting their ships with his daughter and his wife. The
wish is granted ; Hild is cunningly separated from her father. Then the
armed men spring up : anchors are weighed, sails hoisted and the ad
venturers return to Hotel's land ; whilst Hagen, whose ships are unready,
is for the moment unable to pursue.
Wate and his companions, so soon as they land, send news of their
success to Hetel, who hastens to meet his bride : but Hagen soon arrives
in pursuit, and a battle takes place upon the shore. Hagen wounds Hetel,
but is himself wounded and hard pressed by Wate : at last Hetel, at the
request of Hild, parts the combatants. Wate, skilled in leechcraft, heals
the wounded, and Hagen returns to Ireland, satisfied that his daughter is
worthily wedded.
The subsequent story of Kudrun, daughter of Hild and
Hetel, does not concern us ; in many respects it is a duplicate of
the story of the wooing of Hild.
A third version of the story is given by Saxo Grammaticus.
The inferiority of the bombastic Latin of Saxo, that "very
curious Northman," to the perfect Icelandic prose of Snorri, has
caused less attention to be given to Saxo's story than it deserves.
Chronologically it is the earliest of the versions, and it has fine
features of its own.
Hithinus, prince of a Norwegian tribe, loved Hilda, a maiden of
excellent renown, daughter of Hoginus, a chieftain of the Jutes. "For
before they had met face to face, the fame of each had fired the other.
But when they had an opportunity of beholding each other, neither was
able to look away, so constant was the love which held their eyes3."
Hoginus betrothed his daughter to Hithinus, and the two chieftains
became sworn brethren : but Hithinus was falsely accused of having
betrayed Hoginus' daughter. Hoginus believed the report and they thrice
met in combat. First Hoginus attacked Hithinus in a pitched battle, but
was repulsed : a second time they fought singly, and Hithinus fell,
1 Panzer, H.G. 222. * Cf. Kudrun, st. 389.
3 Ed. Holder, v, xlviii ; pp. 159-60.
104 Widsith
grievously wounded ; but pitying his beauty and youth his foe spared his
life (for of old it was held shameful to slay a stripling). So Hithinus was
carried back to his ship by his men. After seven years they met again on
the island of Hithinso, and wounded each other mortally. " It is said that
Hilda burned with such love for her husband that by night she raised up,
through spells, the spirits of the slain, to renew the combat."
It seems probable that Saxo is here harmonizing two
distinct accounts. For whilst it is quite intelligible that the
sorceress Hild, as depicted by Snorri or Bragi, with her valkyrie-
like love of slaughter, should renew the combat from day to
day, it is not clear why Saxo's Hilda should do so, from love for
Hithinus. Again, according to Saxo, it was slanderers who
caused the breach between the chiefs : Hithinus is not repre
sented as having forcibly seized his wife, yet Saxo uses a phrase
patre filiam pertinacius reposcente which shows that he knew a
version in which Hithinus carried off Hilda by force. It is even
possible that Saxo, after the manner of rationalizing compilers
of all ages, is harmonizing his two sources by introducing the
same incident twice in different forms ; and that in the second
and third combats we have traces of two versions, in one of
which the combatants depart alive, reconciled, whilst in the
other the battle is renewed everlastingly1. But it may well be
doubted whether any of Saxo's sources had other than a tragic
ending.
Which, if any, of these divergent versions preserves the
original tale, as it was sung in the halls of Germanic chiefs six
centuries before Snorri or Saxo wrote ? Almost all students
have agreed that the eternal fight of the Heodenings, as told
in the Norse versions, is the essential part of the myth2: that
1 Axel Olrik has attempted to disentangle the two versions used by Saxo.
The eternal battle he thinks is peculiar to the Norse version : the loves of
Hithinus and Hilda, the slanderous tongues, and the localization of the last
fatal battle at Hiddenso, peculiar to the Danish tradition. See Sakses Old-
historic, u, 191-6.
2 On few points has so much unanimity been shown: "Ihre im wesent-
lichen urspriinglichste Gestalt bietet Snorri Der ewige Kampf, das endlose
Hjaftningavig ist offenbar der eigentliche Kern des alten Mythus." Symons
in Pauls Grdr.(2) in, 711; cf. 718, "Wate gehb'rt der Sage nicht urspriinglich
an." "In uralter Zeit von den das Meer beriihrenden frankischen Stammen
ausgebildet ist sie...nach dem Norden gewandert und dort wie so vieles andere
in ihrem alten Zusammenhange erhalten geblieben." Koegel, Ltg. i. 1, 170.
"Der Hiaftningavig aber bildet den eigentlichen Kern der Sage." Martin's
Kudrwift) xnv, XLVH. " The oldest form of the story is to be found in the later
Edda." Robertson, History of German Literature (1902), p. 72.
Stories known to Widsith : Tales of the Sea-folk 105
the features which distinguish the High German epic, such as
the help given by Wate1, or the sweet song of Horant2, had
originally no connection with the story.
This unanimity is the more extraordinary in view of the
fact that the Old English allusions to the story, certainly the
oldest by many centuries, seem to incline rather to the German
than to the Norse version. For it is probably not by a mere
coincidence that in Widsith Wada is mentioned in the line
following that devoted to Hagena and Heoden8. The reference
in Deor is unfortunately obscure, but two things are clear:
firstly that Heorrenda was known to the Old English poets as
a mighty singer, just as Horant later was in Germany : secondly
that this Heorrenda was connected with the Heodenings. So
much we can certainly draw from Deor's words:
" — of myself will I say it : of old I was the poet of the Heodenings,
dear to my lord, and Deor was my name. For many a year had I a good
service and a gracious lord : till now Heorrenda, the man cunning in song,
received the land-right, which of old the lord of earls gave to me."
These references seem to show that Wada and Heorrenda
had their part in the Anglian lay, just as they had in the
Middle High German epic. That, in spite of these references,
critics should have insisted that Snorri's well-told little tale
preserved the original version, is a remarkable testimony to the
power of style.
Breaking away from this tradition, Friedrich Panzer4 saw the
germ of the story not in the eternal battle, but in those parts
which previous critics had regarded as excrescences : in the
loves of Hild and Heoden, in the supernatural help given to
the lover by Wade, in the magic song of Heorrenda. Panzer
believed the story to be of the same origin as the widely spread
fairy tale, known to most in the collections of Grimm and
Asbiornsen, but of which as many as seventy-two variants have
been collected, sometimes from sources as remote as the Swahili
tales of Bishop Steere. In this tale a boy, generally a king's
1 Symons, p. 713. 2 Symons, p. 714.
3 Of course it may be a coincidence : but the chances are against such a
coincidence occurring, and we have no right to assume that it did.
« Hilde-Gudrun, 1901.
106 Widsith
son, falls into the power of a supernatural being — the Eisenhans,
Iron John, of Grimm's story, a wild man who dwells at the
bottom of a lake. From him the boy parts at length, having
acquired magic gifts, generally including that of hair of gold.
In rough disguise, and with head covered, he enters the service
of a king, generally as a gardener ; gains the love of the king's
daughter, who alone discovers his golden hair ; and at last weds
her. Thanks to the help of Iron John he repeatedly saves the
king from disaster in war, riding off unknown, and returning to
his humble disguise. But the king is determined to know who
the strange knight is, and he or his men intercept and wound
the hero. The king recognises from the wound that the
despised lover of his daughter is the same as the strange knight,
and all ends well.
Theories which derive ancient epics from modern fairy tales
ought to be received with caution. To collect a number of
stories in which the same motive occurs ; to argue that they are,
in virtue of this motive, all descended from one original ; to
reconstruct this original, selecting an episode here and a figure
there, till the reconstructed original assumes a likeness, more or
less remote, to some ancient epic or romance, to conclude there
fore that this epic is to be regarded as a literary elaboration of
the original fairy tale — all this is a process in which the individual
feeling of the investigator counts for so much that, though he may
convince himself, he can hardly hope to convince others. Panzer
has argued his case with extraordinary conviction, erudition and
plausibility, but he has made few converts to his theory.
It is unlikely that the fairy tale — of the age and origin of
which we must necessarily remain quite uncertain — is the source
of all the numerous mediaeval romances (many of them bearing
only a most remote resemblance) which Panzer would derive
from it. Certain features recur so repeatedly in all popular story
that a collection of seventy-two fairy tales on the one hand,
and half-a-dozen romances on the other, may easily between
them show many common features, without its following that
all the romances are drawn from the prototype of the fairy tale.
Perhaps the strongest evidence against the identification
of the story with the fairy tale lies in the fact that the
Stories known to Widsifh : Tales of the Sea-folk 107
fairy tale ends in a reconciliation. Now all the evidence of
Northern story points to a tragic ending: there is nothing to
show that the story in Old English was not tragic ; whilst the
oldest German reference is to " the battle betwixt Hagen and
Wate, where Hild's father lay dead1." We can understand why
in the later Middle High German epic the fierce old version was
toned down ; but that so many different versions should have
agreed in turning an original reconciliation into a tragedy would
be almost incredible.
But a new light has been thrown by Panzer upon many
features of the story — the daemonic helper, the magic song of
Heorrenda: these will never again be so hastily dismissed as
later additions to the tale.
Above all, the old English allusions receive from Panzer's
study their due place in the history of the development of the
story. There was something paradoxical in the view which
accepted Snorri's thirteenth century tale as the original version,
whilst condemning as secondary Old English references which
could hardly by any possibility be put later than the eighth
century, and were probably much older. Yet earlier critics had,
in spite of Widsith, excluded Wade from the original tale. He
was a mythical figure, a water giant, and had no place in the
warfare of Heoden and Hagena. It is just because he is a water
giant, Panzer urges, that he so obviously belongs to the story.
He is an essential part of the tale, the supernatural, daemonic
figure who helps the hero at his need ; a figure which agrees
admirably with that of the English Wade. Hence the help
given to students by Panzer's monograph can hardly be ex
aggerated, even though his view of the origin of the story has
not in its entirety met with any general acceptance2.
Whilst believing that Panzer has shown Wada to be a
much older figure in the story than has generally been admitted,
1 von einen volcwige hore wir sagen
der uf Widpinwerde gescach,
ddr Hilden vater t6t loch,
inzwischen Hagenen unde Waten.
Lamprecht's Alexander, herausg. Kinzel, Halle, 1885, 1830 (1321). Compare
note, p. 459.
2 So far as I know, Panzer's view that the Goldener-myth is the origin of the
Hild and Heoden story has not been endorsed by any critic. It has been
108 Widsith
we need not therefore identify him with " Iron- John " : for the
helping and protecting spirit is a figure common enough in all
folk-lore, not belonging necessarily to this one particular story1.
In all probability then, features of the Old Anglian lay were
the wooing of the heroine by the magic song of Heorrenda : the
help given to the hero by the terrible Wada: and the fall
of Hagena, if not of both Hagena and Heoden. But we cannot
reconstruct the story in detail with any degree of certainty.
Critics have been influenced by the interest they have felt
in the vernacular versions of Snorri or of the Kudrun poet. Yet
the charm of these, great as it is, is not the same as that of the
old poetry. It is when we turn to Saxo's version that we are
reminded of the great fragments which have come down to
us from the seventh and eighth centuries. The situations are
those of the heroic age. The tragedy of strife between kinsfolk
and sworn brethren, as we find it in Saxo, was probably the
theme of the old poet of Heoden, as it was that of the poets of
the lays of Finnesburh, of Ingeld, of Hildebrand, and, in a
smaller measure, of Waldere.
In Widsith the story is localized among the Haelsingas, the
Island-Rugians, and the Glommas. The last we cannot identify
with any certainty ; the two former tribes almost certainly
dwelt, in the fourth century, on the islands and peninsulas of
what is now the Baltic coast of Germany2. It should be noted
that in Saxo the last fight takes place in the Baltic upon the
island of Hiddenso, off the coast of Riigen. That Saxo's original
was localized among the Hselsingas and the Glommas is
examined by Much (in Herrigs Archiv filr das Studium der Neueren Sprachen,
cvm, 1902, 395-416); E. Martin (in the Deutsche Litteraturzeitung , xxii, 1901,
2327-30) ; B. Symons (in the Litteraturblatt f. germ. u. roman. Philol. xxm,
1902, 321-8) ; Ehrismann (Z.f.d.Ph. xxxvn, 515), and by an anonymous critic
(in the Athenaeum, No. 3849 ; 1901, n, 152-3). All these critics express dissent,
Much and Ehrismann in a reasoned statement of considerable length. But it
is admitted that Panzer's study has thrown a new light upon the poem. "Damit
sind auch nach meinem Urteil einige Sagenziige erst ins rechte Licht geriickt,"
Much, 397; "Auch in dem zweiten Teile hatte...noch vieles hervorgehoben
werden konnen, dem ich entweder unbedenklich oder doch bedingt zustimmen
kann," Symons. Symons instances the treatment of Heorrenda and the dis
cussion of his original relation to Heoden.
1 Of. Ehrismann in Z.f.d.Ph. xxxvn, 521.
2 See notes to 11. 21, 22.
Stories known to Widsith: Tales of the Sea-folk 109
rendered probable by the fact that the Helsingi, and a roving
captain Glomerus1, are used by Saxo to fill up the intervals
of the Hild story in a strange, inconsequential way, as if he
knew that they had a place in the tale, but did not remember
what it was2.
Hence, although it has been usual to regard the Hild story
as belonging to the North Sea3, its earliest home would seem to
have been among the islands of the Baltic. From these islands
its passage to the adjoining Angles was easy, and they must
have brought it with them to England. The evidence that
it was known to the Franks in the sixth century4 is insufficient5 ;
and the theory that it spread to the Franks from Anglian
settlements in the Netherlands6 depends upon the very un
certain evidence for such settlements. The Old English allusions
are by far the earliest reliable documents which we have for the
Hild story : and as such their importance has hardly, in the
past, been fully recognised.
Evidence points to the story having died out early in
England. The earliest charters frequently show the signature
of an Abbot Hagona ; he and the Abbess Hild of Whitby both
belong to the seventh century. After this date place-names
founded upon the Hild story are found surviving, but men and
women7 bearing the name Hagana or Hild do not meet us
till we find the name Hagana reappearing (owing probably to
Danish and Norman immigration) after the conquest. Wade
alone retains his popularity : but, it would seem, quite inde
pendently of the Hild story.
1 This has been noted by Panzer (H.G. 180).
2 By the Helsingi Saxo of course understands the inhabitants of Helsingaland
in Sweden. To justify a connection between Glomerus and the Glommas many
instances could be quoted for the interchange of the names of chiefs and peoples.
This occurs principally in names ending in ing (cf. note to Hundingum, 1. 23)
but we have also Hoftbroddr and Heaftobeardan (cf. Bugge, H.D. 153, and above,
p. 82).
If the Glommas are to be located near the river Glommen (see note to
1. 21) we have a confirmation of the conjecture of Dr Lawrence that Heoden was
a Geat (Mod. Philol. ix, 36, etc.).
3 Cf. Koegel, Ltg. i, 1, 169 ; Binz in P.B.B. xx, 192 ; Symons in Pauls
Grdr.(2) in, 714.
4 Panzer, H.G. 438. 5 Much in Herrigs Archiv, cvni, 415.
9 H.G. 440.
7 See Binz in P.B.B. xx, 192-6.
110 Widsith
Breca, of the Brandings.
Breoca weold Brondingum.
The Breoca, prince of the Brondings, who is mentioned in our
catalogue of kings, is known to us from Beowulf (499 — 606)
where Unferth taunts Beowulf with his unsuccessful swimming
match with Breca, the son of Beanstan1. Beowulf asserts that he
was the better swimmer, and he silences his opponent by personal
abuse; but his explanation, that he could have swum faster
than his rival, but did not choose to do so, seems insufficient.
And Beowulf nowhere says that Breca was defeated or humiliated.
The poet has not made the swimming match very clear. The
heroes are apparently swimming northward. Breca after seven
nights reaches the land of the Heathoraemas (in the neighbour
hood of the modern Christiania). Beowulf swims side by side
with Breca for five nights, till the flood drives them asunder.
He is attacked by the sea monsters, slays them [on the sixth
night] and sees in the morning before him the sea-nesses of the
land of the Finns. This is hardly intelligible if we suppose
Beowulf to have started from his home among the Geatas. The
story seems most naturally to belong to the ancient Angel, from
which both the land of the Finns (Lapps) and of the Heatho
raemas would alike lie in the vague regions of the distant North.
The mention of Breca in Widsith justifies us in believing
that Breca had a history of his own as a swimmer, even though
the Beowulf poet has introduced him only as a foil to his own
hero2. Breca's name, and that of his folk, point distinctly to
some sea story. Brecan is used in Old English in the sense
1 For Beanstan see Kriiger in P.B.B. ix, 573, Bugge in Z.f.d.Ph. iv, 198,
Bugge in P.B.B. xn, 55 ; Zupitza in D.L.Z. vi, 489-90.
Miillenhoff would connect with O.N. bauni, another reference to the sea and
to sea monsters. (Z.f.d.A. , vn, 421 ; Beovulf, 2.) The word bauni has been
questioned (by J. M. Hart in M.L.N. xvin, 118). It will however be found
in Haldorsen's Islandske Lexikon, Copenhagen, 1814: " bauni, hundefisk" (a
dog fish). I owe this reference to Mr S. Blondal, of the Royal Library,
Copenhagen.
s It certainly does not justify us, with Koegel (Ltg. i, 1, 109), in assuming that
there was an old lay of Beowulf and Breca, used by the poet of Beowulf. For,
since in Widsith heroes of the same cycle are, as far as possible, grouped
together, the absence of Beowulf or Beowa here would be difficult to explain,
the more so since (as has been observed by Moller, V.E. 22) the alliteration is so
convenient.
Stories known to Widsiih: Tales of the Sea-folk 111
of ploughing the sea1 : breki* signifies a breaker in Old Norse
poetry ; brandr in Old Norse means the prow of a ship, whilst
branding, branting has for many centuries been in use among
the sailors of the North Sea to signify " a place where the sea
dashes on the rocks3."
We cannot, then, dismiss as irrelevant or accidental this fact
that both prince and people bear a name which is reminiscent
of the sea4. But, on the other hand, the context, both in
Widsith and Beowulf, shows that Breca of the Brandings is
regarded there as the real prince of a real people, not as a
mythological abstraction of the breakers conquered by Beowulf5.
The swimming contest is a favourite episode in Germanic story,
and Brecca is as concrete an antagonist as was Olaf Tryggva-
son, when he dived under water with Kjartan, till " Kjartan
thought he had never come in so tight a place before."
In the tale of Breca and his Brondings we have clearly some
tale of power over the sea ; exactly what it is difficult to say.
It is going too far to see in the story of Beowulf and Breca a
myth of the Gulf Stream flowing North, and overcoming the
colder polar stream6: for we have no reason to think that this
phenomenon had been observed so early7.
1 "brecan ofer bjej>weg," Elene, 244, Andreas, 223, 513, is used of a boat
being dashed over the waves. See Andreas u. Elene, herausg. v. J. Grimm,
Cassel, 1840, pp. 109, 147.
3 The two words come together in the Edda, fellr brattr breke brondum
hcere (Regensmdl, 17). "The steep billow surges high above the prowa" (Symons
u. Gering, i, 314).
3 Compare Dutch branding, Danish branding, breakers, surge. The use of
brandung in the same sense in High German is comparatively recent. See
Kluge, Seemannssprache, p. 141. There seems no reason (with Chadwick, 269)
to connect the Brandings with the Brand of the Northumbrian pedigree.
4 This, and the absence of any geographical data with which we can connect
their names, make it highly probable that the Brondings are a poetical fiction.
Heyne-Schiicking (314) seeks to locate them in Mecklenburg and Pomerania,
and Thorpe (Cod. Ex. 514) in Branno in the Cattegatt : but there is no satis
factory evidence for either theory. This makes their case different, for example,
from that of the Halsingas of 1. 22. Yet it might be argued that they were a
real tribe or family, that the name Brondingas meant originally " men of the
sword " like Helmingas "helm-men," and perhaps originally Scyldingas, " shield-
men" (Chadwick, 284), and that they subsequently came to be connected with
the hero of a sea story through a false etymology.
8 Cf. Panzer, Beowulf, 271 ; Lawrence in the Pub. of the Mod. Lang.
Association of America, xxrv, 262, 269.
6 Moller, V.E. 22 (note) ; cf. Brandl in Pauls Grdr.(Z) n, 1, 992.
7 Heinzel in A.f.d.A. xvi, 266-7.
For a reference to a possible development of the swimming match story
in the Icelandic sagas, see Bugge in P.B.B. xn, 51-5.
112 Widsith
Theodoric the Frank,
faodric weold Froncum.
The Theodric mentioned as ruling over the Franks is to be
identified with Theodoric I, the son of Chlodowech (Clovis),
who ruled over the Salian Franks from 511 to 534, the conqueror
of the Thuringians, and, through his son Theodebert, of Hygelac
the Geat. And Theodric, no less than Hygelac, has his place
in heroic song.
The poeta Saxo who wrote De gestis Caroli magni toward
the end of the ninth century makes mention of popular songs
in praise of Theodric :
Est quoque iam notum, vulgaria carmina magnis
Laudibus eius avos et proavos celebrant,
Pippinos, Carolos, Hludowicos et Theodricos
Et Carlomannos Hlothariosque canuut1.
About a century later we have a note concerning Theo
doric I: Hugo-Theodoricus iste dicitur, id est Francus, quia
olim omnes Frand Hugones vocabantur, a suo quodam duce
Hugone*. Another two hundred years and we reach the early
thirteenth century, the great age of German epic poetry, to
which period belongs the earliest portion of the Middle High
German poem of Hug-Dietrich and Wolf-Dietrich3.
This story is a romantic one : the relation to the historic
events of the sixth century is slight, and became still slighter
as the tale was overlaid, in the days of the decline of the High
German epic, with wild tales of adventure among paynim
kings. In the name Dietrich, and in the fact that the story
turns upon the suspicion of illegitimate birth, upon the jealousy
of brothers, and the loyalty and disloyalty of retainers, we can
see the last traces of traditions of the historic Merovingian
kings, Theodoric and Theodebert.
We must be careful not to read the Middle High German
story into Widsith, which chronologically stands much nearer
1 De gestis, etc. v. 117, in Pertz (fol.), SS. i, p. 268.
2 Quedlinburg Chronicle, in Pertz (fol.), SS. ni, p. 31 ; see Schroder in
Z.f.d.A. XLI, 24 etc.
3 See Miillenhoff, Die Austrasische Dietrichssage, in Z.f.d.A. vi, 435-59,
and the chapter Teoderico e Teodeberto in the Origini of Pio Rajna.
Stories known to Widsith: Tales of the Franks 113
to the days of the historic Theodoric, than to the German epic
of the thirteenth century. We have seen that we are not
justified in assuming that the Theodric and Seafola who are
mentioned in 1. 115, as retainers of Eormanric, are the same as
Wolf-Dietrich and the traitor Sabene1. The Theodric men
tioned in 1. 24 is undoubtedly Theodoric the Frank, and there
fore ultimately to be identified with the hero of the Wolf-
Dietrich story : but the stories to which he owes his mention in
Widsith are probably not those known to us (in their later
form) through the Middle High German epic of Wolf -Dietrich,
but rather the much earlier traditions, at least half historical,
of Theodoric's conquest of the Thuringians.
This event made so strong an impression upon the tribes of
North Germany as to form a landmark in their annals. Thus
the continental Saxons, according to a lost work of Einhard,
quoted by Adam of Bremen, dated their origin from the period
quo Theodericus, rex Francorum, contra Hirminfridum, ducem
Thuringorum, generum suum, dimicans, terram eorum crudeliter*
ferro vastavit et igne*. The Saxons according to this story
helped Theodoric, and received in return settlements in the
old Thuringian land. In the fragment known as the de
Suevorum origine3, the same story is told of the " Suevi " ;
and the whole legend of Theodoric's Thuringian war, as it was
current in the second half of the tenth century, is given by
Widukind4.
This story is quite dissimilar from the legends later asso
ciated with the names of Hug-Dietrich and Wolf-Dietrich, but
for one remarkable point of likeness. The true hero of the
Middle High German epic is neither Hug-Dietrich nor Wolf-
Dietrich, but the faithful retainer Berhtunc von Meran : to him
is opposed the faithless Sabene. Similarly Widukind's story
1 See above, pp. 41-4.
2 Adam of Bremen, i, 4 in Pertz (fol.) SS. vn, 285 (1846). The same account
occurs, almost word for word, though without mention of its being taken from
Einhard, in the ninth century Tramlatio S. Alexandri. Pertz (fol.) SS. n, 674-5
(1829).
3 Published, with commentary, by Mullenhoff in the Z.f.d.A. xvii, 1874,
pp. 57-71.
4 i, 9-13, in Pertz (fol.) SS. ni, 420-424. The version given below is
a paraphrase and summary of Widukind's story, which is told at too great
length to be given literally.
o. 8
114 Widsith
contrasts the figures of the true and the false servant. Theodo-
ric owes his success, Irminfrid his defeat and death, each to
the counsel of his trusted servant.
Huga, king of the Franks, died, leaving no heir save Amalberga, wedded
to Irminfrid, king of the Thuringians. The Franks anointed king Huga's
bastard son Thiadric, and he at once sent a message to Irminfrid offering
peace and friendship. Irminfrid would have accepted Thiadric's proffer,
had not his wife, claiming that she was heir to the thione, persuaded the
king's counsellor Iring acer ingenio, acutus consilio, facilis ad suadendwm
quae vellet, to advise war.
So, against the will of his chief nobles, Irminfrid replied that Thiadric
was by birth his bondsman, and that he would not yield to him. " Rather
would I give thee my head," replied the messenger, "than hear such
words, knowing that they must be washed out by the blood of many
Franks and Thuringians."
So Thiadric and Irminfrid met ; and the battle lasted three days :
Irminfrid was at last beaten off, though with such loss to Thiadric that
retreat was advised. But, by the words of one counsellor, s&rvus satis
ingeniosus, cujus consilium expertus est saepius probum, eique propterea
quadam familiaritate conjunctus, Thiadric was urged to press his beaten
foe, which he did, summoning the Saxons to his aid. [The story of how
the Saxons helped the Franks does not concern us : in the end Irminfrid
and his Thuringians were utterly routed.] Iring, sent by Irminfrid to beg
for mercy, was persuaded by Thiadric to turn traitor : to lure his master
into the presence of the Frankish king, and then to slay him, in the
moment of making his obeisance. But the promised reward did not
follow. " Hateful to all men for having slain thy lord," cried Thiadric,
" depart from us : we will have neither part nor lot in thy wickedness."
" Justly am I hateful to all men," Iring replied, " having yielded to thy
wiles : but I will purge my sin and avenge my lord." With his still
unsheathed sword he smote Thiadric, and, having placed the dead body
of his master above that of the Frankish king, so that he might conquer
even in death, he cleared a way for himself, and departed. If any faith is
to be placed in this story, Widukind says, the reader must judge, but the
Milky way to this day is known by Iring's name1.
The details of this story are, of course, fabulous: but it
reflects accurately the character of Theodoric, as he is depicted
by Gregory of Tours2, as well as the fact that Irminfrid was
slain, after his surrender, apparently through the treachery of
Theodoric.
Now both the treacherous Iring and the nameless faithful
counsellor seem to belong to the class of retainer known in Old
1 In the Quedlinburg Annals another version of the story is given, in which
Iring is apparently the faithful companion of Irminfrid. See Pertz (fol.)
SS. m, 32. In the de Suevorum origine Irminfrid escapes to the court of
Attila. A version similar to these must lie behind the mention in the
Nibelungen Lied of the two heroes Iring and Irnfrit at Etzel's court. As to the
origin of the Quedlinburg version, see Z.f.d.A. XLI, 24.
2 Hist. Franc, in, 7, 8.
Stories knowm to Widsith: Tales of the Lombards 115
English society as thyle : the professional orator and counsellor1.
It is therefore remarkable, as Miillenhoff noted long ago2, that
in our list Thyle of the Rondings is coupled with Theodric of
the Franks. Thyle as a proper name is in any case strange
enough : can we interpret it as referring to the faithful coun
sellor of the Thuringian war ?
It is difficult to track this story in England. Here personal
names help us but little. The name Theodric is inconclusive,
for Englishmen bearing it were probably named after the
Gothic rather than the Frankish king3 : the faithful servant is
anonymous : no argument can be drawn from the rarity of the
name Iring in England, for a man would hardly name his son
from the villain of the story: the scene of the tale is in
Thuringia, so we should not expect to find it localised in
English place names. It seems clear that, in the name Iringaes
uueg, applied to the Milky way4, we have not necessarily a
reminiscence of the traitor but rather of an ancient mytho
logical figure whose relation to his namesake in the Thuringian
story, and in the Nibelungen Lied, is a difficult problem5, lying
outside the scope of this study.
The Lombards.
Tacitus describes the Longobardi as a small but excessively
warlike nation6, successfully maintaining their independence
against overwhelming odds. This character they retain through
out the ages, so that the sentence of Tacitus would form an
1 For the functions of the thyle, cf. Miillenhoff, D.A. v, 288-9, and Kaufmann
in Sievers' Philologische Studien, Halle, 1896, pp. 159-162. Kaufmann would
make Widsith a thyle, by reason of his relation to queen Ealhhild, and his extra
ordinary knowledge of princes and peoples. For an Irish parallel to the thyle
see Deutschbein in the Germ. Rom. Monatsschrift, i, 114.
2 Z.f.d.A. xi, 280-1.
3 If indeed we are justified in assuming that they were named after either.
For, as saints' names are often given to children at the present day, without
any thought of the patron saint, so a child may have been named Theodric
without thought of any hero : perhaps because he had two uncles, Theodred
and Godric, and his father wished to compliment both at once.
4 Epinal Gloss., ed. Sweet, E.E.T.S. (1883), p. 28.
8 See Grimm, Mythology, tr. Stallybrass, r, 234-5 ; 358-362 ; Symons in
Pauls Grdr. d) n, 1, 32 ; Koegel, Ltg. i, 1, 124-129 ; Grimm, Heldewage, 401 ;
Koegel in P.B.B. xvi, 504.
6 Germania, XL.
8—2
116 Widsith
appropriate motto for the narrative of their national historian,
Paul the Deacon. It is this picture of a valiant little nation, in
the midst of overbearing neighbours, which gives its lasting
interest to their story, from the traditional days of the fight of
Agio and Ibor, and their scanty array1, against the Vandals, to
those of their historic victories over the Herul and the Gepid2.
Traces of the early Lombard home on the lower Elbe
probably remain in the mediaeval Bardengau and Bardowyk3.
It is here too, between the Angles on the North, and the main
Suevic confederation on the South, that all the early authori
ties, Strabo4, Tacitus8 and Ptolemy6, seem to agree in placing
them.
But they must have left this home at some period prior to
A.D. 165, when we find them on the Roman border. Their
departure would leave the Anglian and Suevic folk immediate
neighbours, and there was probably some trouble before boun
daries could be readjusted. Finally, as we know from Widsith,
the matter was settled by the victorious sword of Offa :
heoldon forS si]>>an
Engle and Swaefe, swa hit Ofia geslog.
Like their neighbours, the Angles and Saxons, the Lombards
were less civilized than those East Germanic peoples with whom
the Romans were more familiar. At the beginning of their
career, a Roman historian, who had seen them at close quarters,
characterizes them as ' savage beyond the measure of even
German ferocity V whilst six centuries later Gregory the Great,
whatever fallacies he might cherish as to the angelic nature of
their far-off Anglian kinsmen, could only describe his Lombard
neighbours as ' unspeakable8.'
The persistency of these national characteristics is the more
1 Paulas, i, 7. 2 Procopius (Bell. Gott. ii, 14 ; iii, 34).
3 Of. Hodgkin's Italy, v, 100 ; Von Stoltzenberg-Luttmersen, Die Spuren
der Longobarden, Hannover, 1889.
4 vii, 290 : ^yiffrov pev ofo> rb TWV SoiJjSwp Hdvos- Siijm y&p dird rov 'Pr/vov
M^xpi rov "AAjSios • /u.fyoj 8£ ri aiiruv Ka.1 vipav rou *AX/3ioj i^terat Ka.66.irep ''Ep/j.6v-
Sopoi KO.I AayK6fiap$oi. Strabonis Geographica, rec. Meineke, Lipsise (Teubner),
1852, p. 399.
6 Germania, xxxix, XL. 8 n, 11, 8.
7 Vellejtts, n, 106 (ed. Ellis).
8 See Gregorii i, Papse, Registrum epistolamm, ed. P. Ewald and L. M. Hart-
mann, in M.G.H. (Ep. vn, 23; v, 38; App. n.)
Stories known to Widsith: Tales of the Lombards 117
noteworthy as the Lombards, though frequently mentioned in
the first century of our era, disappear altogether in the middle
of the second, and do not emerge again till the sixth, when
they achieve a number of victories culminating in the conquest
of Italy. As our poem reflects the historic events of the fourth
and fifth centuries, precisely the period when the Lombard
nation was out of the sight of Roman chroniclers, its references
to Lombard kings need peculiar care, and are of peculiar
importance.
Sceafa.
Sceafa [weold] Longbeardum.
No name of any such Lombard king Sceaf is known to the
historians1 ; but it is with this same king Sceaf2 that the
pedigree of the West-Saxon kings began, until a monkish
annalist, by the discovery that Sceaf was a son born to Noe in
the ark, enabled us to trace back to Adam the genealogy
of king George V.
This reference to king Sheaf ruling the Lombards means,
then, that our poet knew of Lombard lays dealing with Sheaf8,
or, if this is saying too much, at least that some allusion
founded upon such traditions had reached him4.
Sceaf is then, apparently, the mythical civilizer whom more
than one of the tribes dwelling near the north sea coast re
garded as their founder, and the parent of their royal race : the
divine foundling who introduces, amid a hitherto barbarous
people, the tillage of the earth5, and with it the settled rule of
a king. The fine story of Sceaf is, by a fortunate chance,
preserved to us, by two Anglo-Latin historians, respectively of
the tenth and twelfth centuries, Ethelwerd and William of
Malmesbury. Taking their account in connection with the
Prologue of Beowulf, we find the story to have been this.
1 Cf. Hodgkin's Italy, v, 176.
2 Sceaf stands to Sceafa as Scyld to Sceldwa or Hreftel to Hreftla. See note
to 1. 32.
3 Koegel, Ltg. i, 1, 104.
4 Cf. Miillenhoff, Beovulf, 10 ; Grimm, Mythologie, 1835, 219.
5 ' So lasst sich nicht verkennen dass hier ein mythus von dem anfang und
der eiufuhrung der altdeutschen cultur vorliegt' — Miillenhoff in Z.f.d.A.
vn, 413.
118 Widsith
He was driven, as a small boy, in a swift boat without any
oarsman to the Island of Scandza1, and found, by the folk of that
land, sleeping, with his head upon a sheaf of corn, surrounded
by weapons. These folk, marvelling, fed and reared the un
known child, till, in his manhood, he came to rule over the
ancient land of the Angles in Sleswig. When he died, full
of years, his people placed him, as he had commanded them, in
a boat filled with weapons and treasures, and let the sea bear
him away, alone, as it had borne him to them2.
The mention of the sheaf of corn, and the connection of
Sceaf with the old Anglian home in Sleswig, we owe to William
of Malmesbury alone. The account of the funeral rites is found
only in the Prologue to Beowulf*.
But here, at once, we are faced with a difficulty. In
Beowulf the hero is called not Sceaf, but Scyld Scefing " Shield
with the Sheaf," and is identified with " Shield," the ancestor-
king of the Danes.
It has been argued that in this Beowulf is true to the
original form of the story: that "Shield with the Sheaf" was
misunderstood as " Shield son of Sheaf " ; that a new ancestor-
king, Sheaf, was thus placed at the head of the genealogy ; and
that the story (which must of necessity be told of the earliest
ancestor) was then transferred from Shield to his supposed
father Sheaf.
1 Not Angeln or Sleswig, as sometimes stated on the high authority of
Kemble (Beowulf, ii, Appendix to Preface, iii) and Grimm (Mythologie, 1835,
218). The point is of some importance.
2 Sceldius [fuit filius] Sceaf. Iste, ut ferunt, in quandam insulam Ger
manise Scandzam, de qua Jordanes, historiographus Gothorum, loquitur,
appulsus navi sine remige, puerulus, posito ad caput frumenti manipulo,
dormiens, ideoque Sceaf nuncupatus, ab hominibus regionis illius pro miraculo
exceptus et sedulo nutritus : adulta setate regnavit in oppido quod tune
Slaswic, nunc vero Haithebi appellatur. Est autem regio ilia Anglia vetus
dicta....
William of Malmesbury, De Gestis Regum Anglorum, Lib. ir, § 116, vol. i,
p. 121, ed. Stubbs, 1887.
Ipse Scef cum uno dromone advectus est in insula Oceani, quse dicitur
Scani, armis circundatus, eratque valde recens puer, & ab incolis illius terrae
ignotus ; attamen ab eis suscipitur, & ut familiarem diligenti animo eum
custodierunt, & post in regem eligunt.
Ethelwerdus, iii, 3 in Savile's Rerum Anglicarum Scriptores post Bedam,
Francofurti, 1601, p. 842.
3 See discussion of Scyld in Olrik's Danmarks Heltedigtning, 1903, pp. 223-
277: Skjold.
Stories known to Widsith: Tales of the Lombards 119
This view, which was originally suggested by Leo1, has of
more recent years become the accepted one ; and the theory
that Scyld is the true hero of the story has been endorsed
by Mo'ller1, Sarrazin8, Binz4, Sievers5, Olrik6, Symons7, Mogk8,
Koch9, Chadwick10, Lawrence", Gering13, Heyne-Schiicking13,
and Henry Bradley14. On the other hand, the older view, that
Sceaf is the original subject of the rudderless boat tale, and
that this story is, in the Prologue to Beowulf, erroneously
applied to the hero Scyld, has been supported among others
by Grimm18, Kemble16, Ettmiiller17, Miillenhoff18, ten Brink19,
Koegel20, Heinzel21, Henning22; and there is more to be said
for it than has of late been generally admitted.
For Sceaf as an ancient king is vouched for
(1) by this passage in the catalogue of kings in Widsith.
(2) by the West-Saxon genealogy in the Chronicle. It is
true that the name does not occur in the oldest MS (the Parker
MS of the late 9th century) and that the three MSS in which it
does occur are all of the llth century. But the name of Sceaf
has simply fallen out of the Parker MS by a scribe's blunder,
together with a group of other names ; and the agreement
of the other three MSS, though each individually is late,
carries us back to the archetype of all three, which is also the
original of the Parker MS itself. We can be certain, then,
that Sceaf was recognized as standing at the head of the West
I Ueber Beowulf, Halle, 1839, p. 24. Grundtvig at an earlier date had
expressed the belief that thie story originally belonged to Skjold (Danne-Virke,
n, 218), but Leo was the first to explain how Sceaf could have supplanted Scyld.
8 V.E. 43.
» Engl. Stud, xvi, 73-80 (Review of Mullenhoffs Beovulf) ; cf. Anglia,
xix, 383 ; xxvin, 410-13. Since writing the above, I notice that Sarrazin
quite recently gave the weight of his authority to the support of Sceaf (Engl.
Stud. XLI, 10).
* P.B.B. xx, 147-150. B S.B. Sachs. Akad. 1895, 176.
6 Helte-Digtning, 223-277. 7 Pauls Ordr.(z) m, 645.
8 Pauls Grdr.Q in, 320.
9 Ynglingar, in Historisk Tidskrift, 1895, 164. 10 282.
II Pub. of the Mod. Lang. Association of America, xxiv, 1909, p. 259.
12 100. 13 320. 14 Encyclopedia Britannica, sub voce Beowulf.
15 Mythology, trans. Stallybrass, 1880, i, 170.
18 Beowulf, u, Postscript to Preface v. 17 Beowulf, 1840, p. 6.
18 Z.f.d.A. vn, 412-14; Beovulf, 10-12.
19 Pauls Grdr.(i)u, 1, 532; Beowulf, 195.
20 Ltg. i, 1, 105 ; Z.f.d.A. xxxvn, 274. D A.f.d.A. xvi, 267-8.
28 Z.f.d.A. XLI, 156-169. Golther (209) and Meyer (193) also give a tacit
support to Sceaf, by telling the story of him, not of Scyld.
120 Widsith
Saxon pedigree in an official list which was drawn up, certainly
before the end of the ninth century, and probably in the
middle of that century1. We have, of course, no allusion in
this list to the " mysterious boat " story : but the fact that
Sceaf is an ancestor of Scyld and stands at the head of the
pedigree, proves that at this date the story must have been
told of him, rather than of Scyld.
(3) We have the evidence of Ethelwerd for the Sceaf
story in the tenth century and
(4) of William of Malmesbury in the twelfth. Late as
William is, the character of the additional details which he
gives shows them to be genuine folk lore, not a mere amplifica
tion of Ethelwerd's story ; and he is a historian who avowedly
drew very largely from popular songs2.
The only evidence on the other side is that already men
tioned of the prologue to Beowulf, where the story told of
Sceaf by Ethelwerd and William is attributed to Scyld.
That Scyld was an early king of the Danes we know from
Saxo: another account makes him primus inhabitator Germanie:
but nowhere save in Beowulf is the story of the boat told of him.
It is quite conceivable that the author of the prologue to
Beowulf has transferred to Scyld, whom he knew as ancestor of
the Danish house, a story which belongs of right to Sceaf, the
ancestor of the West-Saxon house. This would be quite in
accordance with that blending of English and Danish tradition
which we seem to be able to trace elsewhere in Beowulf,
Either theory is possible, and if we had only Ethelwerd and
William on the one hand, and the prologue to Beowulf on the
other, it would be difficult to decide. But the mention of
Sceaf at the head of the pedigree in the Chronicle and still
more this mention in Widsith render very difficult the position
1 I hope later to give a critical text of the West Saxon pedigree, with a fuller
discussion of the Sceaf-Scyld problem.
2 Cf. Gesta Regum Anglorum, n, §§ 138, 148, 188 (ed. Stubbs, vol. i, pp. 155,
165, 230). Macaulay, in his Introduction to the Lays of Ancient Rome, instances
William as a typical example of the historian who draws upon popular song :
Freeman (Historical Essays, First Series) has shown the same. A further
piece of evidence tending to show that William is recording genuine tradition
has been pointed out by Chad wick (278).
Stories known to Widsith: Tales of the Lombards 121
of those who suppose him to be a late creation, a figure formed
from a misunderstanding of the epithet Scefing applied to the
hero Scyld.
Till more evidence appears, it would seem that we are
justified in following in the steps of Kemble and Grimm :
in accepting the story of William of Malmesbury as a true
piece of folk lore, and in accepting Sheaf as the hero of
the story, as the primitive king alike of Lombard and English
tradition.
jfigelmund,
No mention of either Shield or Sheaf is to be found in the
Lombard historian, Paul the Deacon, though he, like William
of Malmesbury, has drawn liberally enough upon the songs of
his people, either directly, or through an intermediate Latin
document, the Origo gentis Langobardorum of the end of the
seventh century. In one case indeed the Origo and, following
it, Paul, have given the old story in such detail that students
have thought that they could trace the alliteration of the old
Teutonic verse under the Latin paraphrase1. This story, telling
how the Long-beards came by their name, is the one which has
been put back into stirring English verse by Kingsley, as Wulfs
song in Hypatia.
After the death of their two dukes Agio and Ibor (the
victors in this battle with the Vandals) the Lombards, we
are told by the Origo and by Paul, were governed by kings, of
whom Agelmund, the son of Agio, was the first. An ^Egel-
mund is mentioned in our poem, together with a hero who
bears the name of another Lombard king, Eadwine (i.e. Audoin,
the father of Alboin).
...Hlij>e and Incgenj>eow
Eadwine sohte ic and Elsan, ^Egelmund and Hungar,
and >a wloncan gedryht Wij>-Myrginga.
It is usually assumed that we have here a series of Lombard
kings, who have been attracted into the circle of Ermanaric,
among whose champions they are numbered. Hlithe has
1 Koegel, Ltg. i, 1, 107-8. Bruckner, Die Sprache der Langobarden (Q.F.),
1895, pp. 19, 20.
122 Widsith
generally been identified with Leth or Lethuc, an early Lom
bard king1 : but we have seen above that the linking of Hlithe
with Incgentheow makes it almost certain that he is rather a
hero of the wars of Goth and Hun2. The evidence which has
led to Elsa being declared a Lombard chief is ludicrously
inadequate3: Hungar, if we can identify him at all, is a Goth8.
The explanation of the Wty-Myrgingas as neighbours of the
Myrgings, i.e. Lombards, depends upon the assumption, which
itself is groundless, that in our poem Myrgings and Lombards
are regarded as allies4.
We are left then with two names alone, Eadwine and
jEgelmund, to support the theory that a group of Lombard
kings and heroes has deliberately been incorporated into the
company of the retainers of Eormanric.
That the Lombard dynasty was connected by the Old
English poets with Gothic story is possible enough. A similar
connection is said to be found in Hungarian legend 5. We have
seen, too, how in our poem Ealhhild, the wife of Eormanric, is
spoken of as daughter of Eadwine6, who is apparently the
Lombard king. But this hardly supports the theory that the
Eadwine mentioned among the household of Eormanric is
identical with the Lombard. For if Widsith escorts Ealhhild,
daughter of the Lombard king Eadwine, to the court of Eor
manric the Goth, then the Eadwine whom he meets there as
one among the Gothic king's retainers can hardly be the
Lombard.
It is quite possible then, but it seems to me unlikely, that
the poet meant to include Lombard champions among the host
of Eormanric, waging battle with the people of Attila.
I should be much rather inclined to surmise that the two
lines containing the names Eadwine, Elsa, jEgelmund, Hungar
and Wfy-Myrgingas are a random interpolation, the work of
the same unintelligent person who inserted the Biblical and
Oriental names among the list of peoples7. Eadwine is mere
repetition, like the repetition of Creacas, Finnas and Casere ;
1 See note to 1. 116. 2 See above, p. 47. 3 See note to 1. 117.
4 See note to 1. 118. B See Matthaei in Z.f.d.A. XLVI, 3 etc.
8 See above, pp. 23-24. * See above, pp. 7-8.
Stories Tmown to Widsith : Tales of the Lombards 123
Elsa may very possibly, like Hwala1, come from the widely
known genealogy of the West-Saxon kings; Wty-Myrgingas
looks like a mechanical repetition of wty Myrgingum in 1. 42,
which the interpolator perhaps took to be one name. He had
a particular weakness for scattering the Myrging name over
the page, cf. 11. 84-5 : mid Myrgingum... ongend Myrgingum.
But what most of all points to these five names being an
interpolation, is that they interrupt the series of heroes of the
wars of Goth and Hun. When we eliminate the interpolation,
the list runs, as it should : Hlithe, Incgentheow, Wulfhere,
Wyrmhere, and then the reference to the fight in the Vistula
Wood.
If these lines are an interpolation, it is not clear how the
interpolator got the name of ^Egelmund. In this form the
name is hardly known in England. It may be a last echo of
the story of the Lombard hero : if it be we cannot judge how
much of his legend was still remembered.
Paul2 tells us how Agelmund saved from the water a child,
whose father was unknown, and caused him to be carefully
brought up : and how this child, named Lamissio3 by the king,
grew in time to be a great champion.
This story is generally supposed to be a reflection of the
Sceaf myth4, but the likeness is not sufficiently strong to
enable us to speak with certainty.
/Elfwine and Eadwine.
Swylce ic ws&s on Eatule mid j?Elfwine...
...beam Eadwines.
The identification of Hlithe and Elsa with Lombard kings
and chiefs is then improbable : but as to Eadwine and his
son ^Elfwine there is no doubt that they are the Audoin and
Alboin under whom the Lombard people emerge again into
i See note to 1. 14. 2 i, 15.
3 Laiamicho in the Origo : which however knows nothing about the marvel
lous origin of the hero, but makes him ex genere Gugingus, like his (foster-)
father.
4 See Leo, Beovulf, p. 31 ; Leo connected the stories of Sceaf, Lamissio, and
the Knight of the Swan ; cf. also Miillenhoff, Beovulf, 10 and (to the contrary)
Koegel, ~Ltg. i, 1, 106.
124 Widsith
the light of history1. Not but what the stories of these two
chiefs, as given by Paul, show clear traces of the Germanic
poet at work : for Paul, writing after the Lombard overthrow,
is separated by more than two centuries from Alboin's invasion
of Italy : probably the poet of Widsith stands nearer to that
event than does the historian of the Lombards. Thus the
greatest of all Germanic tales is woven in Paul's history round
the figures of Audoin and his son Alboin. The story, which
reminds us of the twenty-fourth Iliad by its bringing into
conflict the two strongest motives of the primitive warrior, the
duty of hospitality, and the duty of revenge, tells how Alboin
won a battle over the Gepidae and slew in single combat
Thurismod, the son of their king Thurisind. Yet Audoin,
according to the Lombard custom, would not give his son the
rank of companion at his table, because he had not received
weapons from the king of some other people2.
When Alboin heard this reply of his father, he took with him forty
youths only, and betook himself to his old foe Thurisind, king of the
Gepidae, and told him why he had come. Thurisind received him gently,
invited him to the feast and placed him at his right hand, where Thurismod,
his dead son, had been wont to sit. And, whilst they feasted, Thurisind,
beholding his son's seat, and calling to mind his death, and seeing his
slayer seated in his room, sighed deeply ; nor could he forbear but
that his grief burst forth, and he said " Lovely to me is that seat, but he
who sits in it is hard enough to look upon." Then the king's second son,
moved by his father's speech, began to revile the Lombards, comparing
them, becaxise they wore white cross-garters, to white-shinned mares3.
Then a Lombard replied, " Go to Asfeld : and there canst thou well find
how hard those whom thou namest mares can kick : where the bones of
thy brother lie scattered like those of a hack in the midst of the fields."
The Gepidae were unable to bear this, and, moved to mighty wrath,
pressed to avenge these open insults : the Lombards, on their side, ready
to fight, placed their hands upon their hilts. Then the king, leaping from
the table, threw himself between them and restrained his men : threaten
ing to punish first him who first began the fight. It was no victory
pleasing to God, he said, when a man in his own house destroyed his
guest. So at last the quarrel was stayed, and they finished the banquet
with revelry. Thurisind, taking the arms of his son Thurismod, gave
them to Alboin, and sent him back in peace, safe to his father's land*.
1 The names correspond exactly, and the mention of Italy makes the
identification certain.
2 Similarly Beowulf, even after he has slain Grendel, still sits with the
giogffS, and with Hrothgar's two young sons. — Beowulf, 1189-91.
3 The point of Fetilce sunt equee quas similatis is not clear. Waitz supposes
Fetil(S=faetulce. Sievers (P.B.B. xvi, 363-5) and Koegel (Ltg. i, 1, 118)
connect it with O.H.G.fizzil, dappled. But is this sufficiently offensive ?
* i, 24.
Stories known to Widsith : Tales of the Lombards 125
What element of truth this story of Alboin may contain, we
can never know : the events in the background are historical
and certain enough. The great battle between Lombards and
Gepidae probably took place in 554. In 565 Audoin died, and
Alboin became king: he annihilated his ancient guests in a
second great battle in 567 : and on Easter Monday, 568, he set
out from the highlands of Noricum, which had long been the
Lombard home, to conquer Italy.
But the Lombards had always been a scanty people, and
had lost heavily in the terrible battle in which Alboin had
struck down Thurismod1. Not merely 'Widsith/ but all the
masterless men in Germany, would seem to have joined Alboin's
ranks. Two hundred years later, the descendants of Gepidae
and Bulgarians, Sarmatians and Suevi could still be distin
guished in Italy2. Some thirty thousand Saxons, with their
wives and children, joined the invaders.
Wasted as Italy had been by pestilence and famine, it
could make no adequate resistance to this motley host, and,
with the exception of the great coast towns, Ravenna, Rome,
and Naples, the whole country was soon in the hands of the
conquerors : not, however, before Alboin had been murdered :
according to the Lombard story by his wife, a Gepid princess,
in vengeance for the wrongs which he had inflicted upon her
race.
Perhaps Alboin, like Attila, possessed exceptional powers of
holding together jarring nationalities : certainly his death, like
that of the Hun, was followed by dissensions among his host.
The Saxons were given the choice of adopting Lombard laws
and customs, or leaving the country : and they chose to go8.
Most of them were slaughtered in their attempt to regain their
old homes. But many of Alboin's host, disappointed, in the
anarchy which followed his death, of their share in the plunder,
1 Jordanes refers to the battle as one of the three bloodiest of recent times,
cecideruntque ex utraque parte amplius LX milia. Jordanes, Romana, ed.
Mommsen, p. 52.
2 Certum est autem, tune Alboin multos secum ex diversis, quas vel alii reges
vel ipse ceperat, gentibus ad Italiam adduxisse. Unde usque hodie eorum in
quibus habitant vicos Gepidos, Vulgares, Sarmatas, Pannonios, Suavos, Noricos
tive aliis huiuscemodi nominibus appellamus. Paulus, Hist. Lang, n, 26.
3 Gregory of Tours, Hist. Franc, rv, 42 ; v, 15. Paulus, Hist. Lang, n, 6 ;
in, 5-7. Hodgkin's Italy, v, 189.
126 Widsiih
may have returned to their northern tribesmen with tales of
the treasures of Welsh land, and of the great deeds of its
conqueror.
So a very few years after Alboin's death his name may
have been a household word in North Germany : while at the
same time the distance of the scene of his exploits would make
him seem, to the wild North Sea tribesmen, a half-mythical
hero ; since for popular poetry ' a distance of a thousand miles
is as good as one of a thousand years1.' Before the end of the
sixth century the names of ^Elfwine, Eormanric and ^Etla may
well have been thrown together as they are in our poem : nor
are we necessarily compelled to regard the reference to Alboin
as an interpolation differing very widely in date from the
matter of which the bulk of our poem is made up.
That, two centuries later, Alboin was remembered among
High and Low Germans alike we know on the authority of Paul
the Deacon. " His generosity and his glory, his valour and his
success in war, are celebrated in the songs of Bavarians and
Saxons, and other people of kindred speech2."
It is a pleasing accident that our catalogue of heroes, as
it includes Eastgota, the first, should also include ^Elfwine,
the last, the wildest and the most romantic of the German
conquerors of Rome.
1 On pent dire que le respect que Von a pour Us heros augmente a mesure
qu'ils s'&loignent de nous. L'eloignement des pays repare, en quelque sorte,
la trop grande proximite des temps, car le peuple ne met guere de difference entre
ce qui est, sifose ainsi parler, a mille ans de lui, et ce qui en est a mille lieues.
Racine, Preface to Bajazet.
2 Alboin vero ita prceclarum longe lateque no/men percrebuit, ut hactenus etiam
tarn aput Baioariorum gentem quamque et Saxonum, sed et alios eiusdem lingua
homines, eius liberalitas et gloria, bellorumque felicitas et virtus in eorum car-
minibus celebretur. Paulus, Hist. Lang, i, 27.
CHAPTER IV.
WIDSITH AND THE CRITICS.
THE views which scholars have held about Widsith have
mainly been based upon a literary study of its style and heroic
legend. Before going on to examine the geography, grammar
and metre of the poem, we may see what measure of agreement
has been arrived at upon grounds of style and legend. We can
then decide whether or not these conclusions are confirmed by
a study of metrical and grammatical details.
Natural divisions of the poem.
There is, at any rate, a prima facie case for the theory
that Widsith is compounded from several sources. If, as is
likely enough, several metrical catalogues existed in early times,
enumerating kings and tribes, they would be likely to coalesce.
And it cannot be denied that the poem, as we have it, falls into
certain well-defined sections :
(a) First, we have an introduction of nine lines, charac
terizing Widsith, and introducing the poem proper : " Widsith
the far-travelled Myrging, who, with Ealhhild, sought the home
of Ermanaric, spake and said " —
(A) Then follows a catalogue of kings and tribes, two
kings and two tribes being generally mentioned in each line :
^Etla weold Hunum, Eormanric Gotum.
The Traveller's personality is here kept quite in the back
ground. The list, it is true, begins "I have heard of many
kings ruling over the nations " : but ic gefrcegn is an epic
formula with little meaning. Widsith is not said to have
128 Widsith
himself visited either the kings or their folk. After seventeen
lines in which the names of kings and princes are crowded
together, we get, in more detail, allusions to the stories of Offa
of Angel, of Hrothwulf, Hrothgar and Ingeld.
(B) Then the Traveller gives the list of the tribes which
he himself has visited. For the most part these are grouped
three in a line:
Mid Wenlum ic wses, and mid Waernum, and mid Wicingum.
It is among these names that the evidently late passage of
Biblical and Oriental lore occurs. The chiefs alluded to in
section (B) are few, and they are mentioned, in some detail,
with words of praise. Particular honour is done to the lady
Ealhhild. Her praise the poet has sung through many a land.
((7) " Thence I fared through the land of the Goths, and
visited the champions of Ermanaric." A catalogue of these
Gothic champions follows : two, Wudga and Hama, are selected
for special honour.
(6) Finally come nine lines of epilogue. " In such wise do
gleemen wander from land to land." This passage, like the
introductory lines, is not supposed to be spoken by Widsith,
but is the moralization of the poet.
Without prejudice to the question of unity of authorship,
we will call A, B, C, respectively the " Catalogue of kings," the
" Lay of Ealhhild " — for it is to her praise that this portion of
the poem leads up — and " The Champions of Ermanaric."
Early Criticism : Mullenhoff, Holier, Ten Brink.
Yet, in spite of the ease with which it may be divided up,
our poem, with the exception of the one passage in which
the names of the Biblical folk occur, bears upon it the impress,
if not of one mind, at any rate of one race and of one period.
We have seen that this suspicious passage was rejected
quite early : by Kemble in 1833, and, with the addition of
another two lines, by Leo in 1838. Six years later (1844)
came MiillenhofFs study of the poem in the Nordalbingische
Studien. Mullenhoff tried to distinguish a number of early
Widsith and the Critics 129
interpolations, such, for instance, as the passages dealing with
Offa, Hrothgar and Hrothwulf1. He did not, however, make
any detailed dissection of the poem into its presumed com
ponent parts.
In his subsequent study, fifteen years later, in the Zeit-
schrift fur deutsches Alterthum, he abandoned any such dis
section as hopeless. He rejected the lines already rejected by
Kemble and Leo, together with some others, twenty-one in all2,
which he regarded as the work of one and the same inter
polator. " I cannot distinguish," he wrote, " earlier and later
additions in this poem. It appears to me to have undergone
interpolation from one hand only. But let no one think that,
by rejecting these additions, he can restore exactly the original
poem. In its essence it is certainly the oldest which Anglo-
Saxon literature possesses. But when was it first written
down? And has no word been distorted, deranged, or lost3?"
Mullenhoff, then, allowed fully for the different corruptions
which might have crept in between the seventh century and
the eleventh. He also recognized, what could scarcely have
escaped the observation of a much less able critic, that the
poem fell into three main divisions : the catalogue of kings ;
the catalogue of tribes and chiefs whom Widsith had known ;
and the Ermanaric catalogue4. But Miillenhoff at this date
seems to have regarded all these as merely sections of one
poem, the work of one mind8.
Close on a quarter of a century later, in his Beovulf, he
1 Nordalb. St. i, 162. Compare also " Wenu nun aber Burgunden genannt
werden und ihr Konig jetzt Giinther Gudhhere genannt wird, so konnte das
wieder nicht derselbe Sanger than der friiher ihnen Gifica gab " (p. 165).
2 14-17, 75-87, 131-134. 3 Z.f.d.A. xi, 294.
4 Miillenhoff in 1859 divided thus :
1-9. Introduction.
10-49. First section, with 10-13 introduction [rejecting 14-17].
50-108. Second section, with 50-56 introduction [rejecting 75-87, thus
bringing the passage in praise of .(Elfwine nearer to that in praise of his sister
Ealhhild].
109-130. Third section.
131-143. Conclusion [rejecting 131-134].
He adhered to this division in his later work, but added 45-49 to the list
of supposed interpolations.
5 Cf. Z.f.d.A. xi, 285, where unity of authorship is assumed. Some years
earlier (1848) Mullenhoff seems to have held the theory, later advanced by
Moller, that the Ermanaric-catalogue differed in date from the rest of the poem.
In the Zj.d.A. vi, 458 he attributes it to the eighth century.
c. 9
130 Widsith
expressed his agreement with the theory which recognized
in these three sections three originally unconnected poems.
But with regard to details he was still cautious. " The work
of each of these poets can be separated from that of the two
others, but not from that of the man who bound the three lays
into one." All three poems, he thought, bore the stamp of the
same age, or at least fell within a century of each other : the
period was the later sixth and the earlier seventh century :
the first section, the catalogue of kings, was the oldest1.
In regarding the three lays as the work of distinct authors,
Mlillenhoff was following in the steps of Hermann Moller, who,
in his Altenglische Volksepos (1883), had attempted to define the
work of the different poets with that exactitude which Miillen-
hoff regarded as beyond the power of the critic.
Broadly speaking, Moller divided the poem as Mullenhoff
had done in 1859 ; he distinguished a prologue (11. 1-9)
followed by (I.) the catalogue of kings (10-49), in which he
supposed the more detailed passage in praise of Offa and
Hrothgar to be a later addition ; (II.) the Ealhhild lay (50-81,
90-108) ; (III.) the catalogue of Ermanaric's champions (88-90,
110-134); and finally the epilogue (135-143). A number of
minor alterations were made in the poem by Moller, in order
to force it into those four line stanzas in which he believed the
Old English lays to have been originally composed. But the
gist of Moller's criticism lay in the fact that, whilst accepting
Mtillenhoff's division of the poem into three, he confidently
attributed each of these three sections to a different author
and a different period.
One further essential alteration he made. Four persons
are specially praised in the Ealhhild lay for their gene
rosity : Guthhere the Burgundian, JElfwine son of Eadwine
the Lombard, Eormanric the Goth, whose splendid gift to
the poet is specially mentioned. This gift, as we know,
Widsith, on his return home, presented to Eadgils, his lord,
from whom he held his fief. Then, in conclusion, we have the
praise of the generosity of Ealhhild, daughter of Eadwine.
1 Beovulf, 1883 (published posthumously, 1889), p. 92.
Widsith and the Critics 131
But Moller would not allow a place to Eormanric in the
Ealhhild lay. That Eormanric should have bestowed a ring
upon our poet was, he argued, chronologically impossible,
whilst a gift from ^Elfvvine need not have been fictitious.
Secondly, ^Elfwine is praised in the poem as the most generous
of princes. So it must have been he who gave our singer his
most precious gift. Finally, as Moller read the poem, Eadgils
was the husband of Ealhhild, and consequently brother-in-law
of ^Elfwine. " The singer after his return would have pre
sented to his lord the gift of an allied and related prince, not
of any other."
This last argument is certainly unsound. Teutonic custom
allowed the passing on of a gift from one great chief to another,
without postulating any friendship between them. Besides,
we have seen that the supposed relationship between ^Elfwine
and Eadgils is an entirely baseless assumption1. The first
argument is without weight to those of us who hold the whole
of Widsith's travels for an epic fiction. The second argument,
that our poet graduated his scale of praise exactly in accord
ance with the supposed generosity of the chiefs, can hardly be
maintained seriously. Eormanric, "more cunning than all in
guile, more generous in gift2/' could not be praised as enthu
siastically as ^Elfwine, though he might be represented as
giving Widsith an even more magnificent gift.
Yet on these grounds Moller, by transferring certain lines
from one section of the poem to the other, excluded the name
of Eormanric from the Ealhhild lay, and made ^Elfwine the
donor of the precious armlet3.
Holler's dissection of the poem commanded in general, as
we have seen, the assent of Mlillenhoff. It was also followed,
in its main details, by Ten Brink. Individual lines, which
Moller had cast out, because they interfered with his arrange
ment of the poem in strophes, were indeed readmitted by Ten
Brink, who owed no allegiance to this metrical theory. And
in one point particularly Ten Brink greatly improved upon
Moller's scheme. He recognized the close relationship which
1 Se.e above, p. 21 etc. 2 See above, p. 34. 3 V.E. 3, 9.
9—2
132 Widsith
exists between the second and third sections of the poem, as
divided by Moller ; the Ealhhild lay and the Ermanaric cata
logue. So close a connection could only be accounted for,
he thought, by their having been early fused into one poem :
lines had been transferred from the one to the other, and the
two poems had become so wound together that it was difficult
to separate them with any certainty1 ; and they had been so
combined, he held, since the latter half of the seventh century2.
Ten Brink followed Moller in regarding the Offa episode as
a later addition to the catalogue of kings, but one made, per
haps, whilst the Angles were still on the continent3.
One may doubt whether it is possible for any critic, how
ever brilliant, to trace the minute history of a poem so far
back into heathen times. But it should be noted that, in
Ten Brink's opinion, the compiler who, perhaps in the eighth
century, put our poem together in its present form, had before
him only two songs : on the one hand the catalogue of kings
together with the Offa and Hrothgar episodes, and on the other
hand the contaminated lay of Ealhhild-Eormanric. Both of
these, whatever their ultimate origin, Ten Brink believed to
have been complete poems for some generations.
In Ten Brink's view Ealhhild was the wife of Eadgils : her
relation to Eormanric was negligible : she had on one occasion
visited his court. Hence he naturally regarded the Ealhhild
and the Eormanric episodes as distinct in origin, although very
early brought together. Now we have seen that the balance
of probability is strongly in favour of Ealhhild being, not the
wife of Eadgils, but, as suggested by Heinzel4, another form of
Sonhild or Swanhild, the luckless wife of Ermanaric. But,
if Ealhhild is to be identified with Swanhild, who has no
existence in story apart from Ermanaric, it is absurd to con
tinue to speak of an Ealhhild lay and an Eormanric lay as
distinct. The second and third sections of the poem must be
one lay. And here comes in the importance of Ten Brink's
1 P. 540. 2 P. 544. 3 P. 539.
4 W.S.B. cxiv, 514-16. See above, p. 22 etc. It is not necessary for the
argument that Ealhhild should be Swanhild, but only that she should be wife
of Eormanric. And so much, at least, seems fairly certain.
Widsith and the Critics 133
conviction, that the Ealhhild lay and the Eormanric lay are not
distinguishable in metre and style. His adherence to the old
view that Ealhhild is Eadgils' wife forced him to regard these
two sections as originally distinct lays. When, in spite of this
presupposition, he recognizes their remarkable likeness1, we
have evidence of the greatest value : for it is that of an able
critic given, not in support of, but in opposition to, his pre
conceived theories. In order to make the phenomena of the
poem fit these theories, and to build up a distinction between the
Ealhhild and Eormanric lays, Ten Brink had to transfer lines
from the one section to the other.
At least then from line 50 to line 130 the poem would seem
to be a fairly consistent whole. It may be compounded of
various elements, and it is probable enough that reminiscences
of a dozen different lays lie behind it. But everything has been
so fused and absorbed that it is impossible for us to distinguish
the different strata, or to subdivide the poem further2. We
cannot follow Ten Brink in venturing back into the remote
past, and showing how the seventh century poet put the
Ealhhild- Eormanric lay together ; we have not evidence enough
to do so3. All we can do is to mark as doubtful any lines like
the 'Biblical' passage, which seem to indicate a date later
than that of the bulk of the poem.
But are we justified in following Ten Brink in maintaining
a distinction between this Ealhhild-Eormanric lay on the one
hand, and the preceding catalogue of kings (11. 10-49) on
the other ?
Here, I think, a further examination entirely bears out Ten
Brink's theory. There is a clear break in the continuity of
the poem at 1. 50 4. Swa, in view of what has preceded, makes
no sense. And the passage which precedes is of a different
character from that which follows.
The catalogue of kings is an impersonal list : the Ealhhild-
Eormanric lay is a personal narrative. They have no organic
1 Pauls Grdr.d) ir, 540-541. Compare for example 11. 68, 69, with U. 113,
114.
2 Always, of course, excepting the late " Biblical" interpolation.
3 For an opinion to the contrary see Schiitte, Oldsagn, 47.
4 See note to that line, and cf. Moller, V.E. 34.
134 Widsith
connection ; to print them as separate poems would not involve
the change of a single word. And there are serious dis
crepancies between them.
Whereas the scop of the Ealhhild-Eormanric lay owns for his
lord Eadgils, prince of the Myrgings, the poet of the catalogue
of kings mentions with pride the victory of Offa over the
Myrgings at Fifeldor. Old Germanic literature is chivalrously
ready to applaud valour, on whichever side displayed : but the
exultant reference to Offa's victory cannot have been intended
to have been put into the mouth of Widsith, who is represented
as belonging to the defeated foemen.
In the catalogue of kings Gifica rules the Burgundians,
Meaca the Myrgings, Sceafa the Lombards. In the Ealhhild-
Eormanric portion Guthhere rules the Burgundians, Eadgils
the Myrgings, Eadwine and JElfwine the Lombards. The
^Elfwine passage has been thought by many critics to be a
later interpolation. Granting that it is, the contrast is still
sufficiently striking between the at least half-historical figures
of Eadgils and Guthhere and the apparently quite mythical
figures of Gifica, Meaca, Sceafa.
It may be said that there is here an intentional distinction.
The first list enumerates mythical kings of old time of whom
the poet has but heard ; the other enumerates those semi-
historical figures who were nearer his own day, whom his
Widsith has actually seen1. But on this hypothesis it would
be difficult to account for the inclusion of JEtla and Eormanric
in both lists.
Again, in the first list the Burgundians are apparently
thought of in their old eastern seat by the Goths ; in the
second, under Guthhere, they must have been on the Rhine.
Sceafa in the first list assumes (perhaps) the old Lombard
home by the sea ; ^Elfwine in the second list rules the Lom
bards in Italy. The reference to the Rugians in the first list
1 Cf. Heinzel, A.f.d.A. x, 232. Heinzel, reviewing Moller's work, refuses to
accept his view of three lays welded together. He admits two lays only : roughly
the catalogue of kings and the champions of Ermanaric. The lay of Ealhhild he
divides between the two. But the effect of this division of the Ealhhild lay is to
separate the praise of ./Elfwine from that of his sister Ealhhild. It is much
easier to take the whole of the " Ealhhild lay " with the " champions of
Ermanaric."
Widsith and the Critics 135
assumes that they are a seafaring folk : in the second list they
are mentioned together with the Romans ; a reminiscence, it
may be, of the days when they ruled in Italy together with the
Ostrogoths. The variance in these two last instances is how
ever too hypothetical to have more than a slight subsidiary
value.
Only one of these discrepancies is in itself very convincing :
but taken together, in the bulk, they seem to indicate a
difference1, both in age and character, such as that which Ten
Brink postulated, between the catalogue of kings and the
Ealhhild-Eormanric lay.
Further, the phenomena of the poem justify the distinction
which Ten Brink made between the catalogue of kings proper,
and the lines celebrating Offa and Hrothgar, which he regarded
as later excrescences upon that list. For in the catalogue
proper one king only is allotted to each tribe : Sigehere ruled
the Danes. When we find a second Danish king Alewih, and
a third and fourth Hrothgar and Hrothwulf, it is manifest that
they interfere with the scheme of the original list, and probable
that they are additions to that list made by a later hand2.
Merits of the older criticism.
The view expressed above that the poem falls into two
divisions, the catalogue of kings, and the Eormanric-Ealhhild
lay, is not inconsistent with the views of the earlier critics,
Miillenhoff, Moller, and Ten Brink. Indeed, it develops out of
these views inevitably, so soon as we abandon the unjustifiable
assumption of these older critics that Ealhhild is the wife
of Eadgils.
In recent years a reaction against the school of Miillenhoff
has set in. So far as this reaction is a protest against the too
1 Of. Schiitte, Oldsagn, 47 ; for arguments to the contrary, Heinzel as above.
2 Mention should perhaps be made of the view, originally suggested by
Ettmuller(2), 211, that Widsith is an episode, a fragment from some longer epic.
This was also held by Wright (Biog. Brit. 1842, i, 4), Thorpe (Cod. Ex. 512-3)
and Klipstein (n, 422). None of them produced any arguments in favour of
this theory : yet they were right in feeling that Widsith is no isolated phe
nomenon, but one fragment which has been saved from a great body of story
which has been lost.
136 Widsith
great certainty sometimes claimed by that school for its results,
it is justified. But, on the whole, a comparison of the methods
and results of Miillenhoff and his contemporaries with those of
some of their critics is in favour of the earlier school. The
results of the older students of Widsith, up to and including
Ten Brink, represent a consistent evolution, resulting from
many years of patient study and interchange of ideas. No
man's work breaks violently away from that of his predecessor.
The views of Kemble, Leo, Miillenhoff in 1844, Miillenhoff
again in 1859, Rieger, Moller, Miillenhoff a third time as
published in 1889, and Ten Brink all develop quite naturally
the one out of the other.
Certainly, in their later stages, these views become some
what complicated. But it is no true objection to the theories
of Miillenhoff, Moller and Ten Brink to say that they are
" complicated." Of course they are, ex hypothesi. These critics
believed that the old English poems had had a complicated
past ; and who will deny that much lies between the defeat of
Chochilaicus and MS Cotton Vitellius A. xv ? A complicated
theory of how Widsith might have arisen may not be con
vincing in every detail. But if it shows us the way in which
the poem might have grown up, it is surely better than a
theory of artificial simplicity which does not correspond to the
complicated and sometimes even contradictory traces of date
and origin which our poem bears.
Recent Criticism of Widsith.
The study on the style and interpretation of Widsith re
cently published by Dr W. W. Lawrence1 is characterized by
this distrust of the methods of the older school.
The elaborate patchwork theory of Ten Brink, who distinguishes in
the piece four separate lays, not including introductory and connecting
material, has never been adequately criticised and refuted, although the
general weakness of his method is apparently coming more and more to
be recognized2....
It is not so difficult for an unprejudiced person to admit that some
such additions as Miillenhoff describes may have crept into Widsith, how
ever unlikely he may think it that Miillenhoff succeeded in denning their
limits with certainty3.
1 In Mod. Philol. iv, 1907, pp. 329 etc.
2 P. 332. » P. 342.
Widsith and the Critics 137
But Mullenhoff did not claim to have defined these limits
with certainty. On the contrary, he frequently declared that
this cannot be done. Miillenhoff, like other critics of the
"dissecting school," was often very definite in his theories:
but the willingness which he showed to reconsider these
theories acquits him of any charge of undue dogmatism. It is
not sufficiently realized that this definiteness of theory is the
only alternative to a vagueness and confused thinking which
must ensue, if we argue about interpolations'without defining
to ourselves their exact scope. It is therefore the duty of
a critic who believes a poem or play to be the work of several
hands, to form a hard and definite theory, consistent with the
facts that he has noted. The law of chance is against his theory
being right in every detail. But a critic who is quite clear in
his own mind as to exactly what he is trying to prove may
often prove his general theory, even though we are in doubt as
to many of the details. If he confines himself to generalities
he will prove nothing.
For example, Mb'ller thought he detected traces of a strophic
system in the Old English epic, and he accordingly published
a Beowulf, arranged in four-line strophes. He cannot have
hoped that his reconstruction reproduced exactly the original
Beowulf. Yet his method was right enough, for the only way
of testing his theory was to attempt such reconstruction, and
to see how it worked out. Most critics will agree that it
works out badly ; that it involves too much violation of the
text. But Moller did a service by demonstrating how much
violation of our existing text is necessary in order to force
Beowulf into strophic form.
Rejecting the theories of the origin of the poem evolved by
Mullenhoff, Moller, and Ten Brink, Dr Lawrence gives a study
of Widsith of which the general principles are indicated by the
following extracts :
If Widsith is inferior in poetic quality to other pieces of lyric character
in Anglo-Saxon, it is by no means wholly lacking in this respect. The
passage describing the singer's relations with his lord Eadgils and with
queen Ealhhild (11. 88 ff.) serves to indicate what the general tone of the
poem in an earlier form may have been. For, as will be seen, closer study
138 Widsith
shows that it has been much overlaid and defaced by the addition of
inferior material, like a Gothic building rudely modernized with bricks
and mortar1....
We shall surely not err in looking to this thread of story for the
earlier material at the basis of the poem, rather than to the lists of names,
etc. which precede. Instances of expanding a tale by the interpolation of
inferior matter are common enough, but to enliven cataloguing by the
composition of epic verse dealing with different material, and telling
a separate story, is, so far as I am aware, unheard of. It seems reason
able, then, to regard much of this ethnological tediousness as a later
addition to the main theme, having crowded out earlier portions of the
poem, so that the real narrative of Widsith's adventures is preserved in a
fragmentary condition only2....
One would like to believe that the references to Guflhere (11. 64-67)
and to ./Elfwine (11. 70-74) formed originally a part of the same story
as 11. 88 ff., as they are similar to it in style and metre, and unlike
the material in which they are imbedded. It is almost impossible to
resist the conclusion that there is here preserved some of the good old
piece which formed the nucleus of the present poem, much mutilated and
interpolated, indeed, but still showing its presence, wherever it remains,
by its superiority to the matter which surrounds it3.
But is not this open to the very objection which Dr Law
rence brings against the school of Mullenhoff : that of " depen
dence upon subjective and a priori conceptions of Anglo-Saxon
style"? Who is to decide what is inferior matter, and what
better material ? And why should the better be necessarily
old, the inferior new ?
Mullenhoff and his school did sometimes, it is true, reject
a passage for no other reason than that it appeared to them
inferior, and therefore, they argued, was later. But, generally,
their criticism depended upon a number of discrepancies, found
in the course of a minute study of details of ethnology, heroic
lore, dialect, and metre. In so far as it depended upon these
things it was more reliable than the criticism which threatens
to replace it, which is based, to a much greater degree, upon
the personal impression of the critic.
Why should the superiority in style or interest of the
passages dealing with the generosity of Guthhere, ^Elfwine,
Eormanric, Eadgils, and Ealhhild prove these passages to con
stitute parts of the original poem ?
And does such superiority exist ? True, a modern reader
will find these parts more interesting than those which are
crowded with names. But this is because he does not know
1 P. 330. 2 P. 340. 8 P. 343.
Widsith and the Critics 139
the stories, and therefore the poet's allusions awake no echo in
his mind. Working on this principle, a modern Bentley might
eject much " inferior matter " from the text of Milton :
Nymphs of Diana's train, and Naiades
With fruits and flowers from Amalthea's horn,
And Ladies of th' Hesperides, that seem'd
Fairer than feign'd of old, or fabl'd since
Of Fairy Damsels met in Forest wide
By Knights of Logres, or of Lyones,
Lancelot or Pelleas, or Pellenore.
Mid Rugum ic wees and mid Glommum is a statement
which, to a modern reader, may be devoid of interest. But
there was a time when the naming of Rugians and Glommas in
the same breath would have sufficed to call up one of the most
glorious of the old tales. It is our misfortune that many of the
lines no longer give up their secrets to us ; it is our fault if we
therefore condemn them. And our poem shows no little art in
the way names are grouped so as to recall the old stories, and
the monotony of the lists is varied by occasional rhyme or
transposition :
peodric weold Froncum, pyle Rondingum,
Breoca Brondingum, Billing Wernum.
Oswine weold Eowum and Ytum Qefwulf,
Fin Folcwalding Fresna cynne.
Sigehere longest Sae-Denum weold,
Hnaef Hocingum, Helm Wulfingum,
Wald Woingum, Wod pyringum, —
I cannot call verse like this 'inferior' — the poet seems to me
to have accomplished a difficult task extraordinarily well. Still
less can these lists of names be condemned as ' late.' On the
contrary, it is in virtue of the proper names, and of these only,
that so many critics have pronounced Widsith the earliest of
all English poems. For these names of kings and tribes show
an extraordinary knowledge of the old stories; Maldon, on
the other hand, shows us how the old commonplaces were
applied to a new generation of heroes, and how descriptions of
the bestowal of rings might have been written during the latest
period of the Old English epic.
Indeed, Dr Lawrence admits that some of the interpolated
passages " may have existed previously, and may be as old as
the narrative portion, or even older." But, if this is so, we are
140 Widsith
driven back to the old theory of the contamination of two or
more lays. I cannot see how the poem can be compared to
" a Gothic building rudely modernized with bricks and mortar "
if the bricks and mortar form a building which is in some cases
as old as or even older than the Gothic building itself. It is
the great merit of the clearly defined theories of Mb'ller and
Ten Brink that, by their very dogmatism, they saved the critic
from the danger of self-contradiction1.
How contradictory recent criticism of the poem has proved,
when compared with the criticism of the school of Miillenhoff,
is shown by the fact that Professor Brandl singles out as the
oldest element in the poem just those lists of names which
Dr Lawrence regards as the modern bricks and mortar2.
The poem, Professor Brandl thinks, may have been put into
its present form in the eighth century. It was constructed out
of three old mnemonic catalogues, which can be distinguished
by their style from the story of Widsith the wanderer, in which
they are framed.
We have seen that there is reason for thinking that the
catalogue of kings is such a mnemonic list, and that it betrays
its origin by certain inconsistencies when compared with the
rest of the poem. In fact, as Ten Brink said, 11. 18-34 are
" uralte versus memoriales."
But that the two other catalogues, of peoples and of
Eormanric's champions, are to be distinguished from each
other, and from the narrative in which each is imbedded, is not
so clear. No doubt there is a difference in style between the
narrative and the lists. How Widsith sang in the hall of
Eadgils is told in the manner which the Old English poets
love : in lines following the one upon the other, with sense
variously drawn out. On the other hand, the names of the
tribes and champions he visited are enumerated in the most
rapid way. But are we justified in assuming that the verse of
an Old English bard was always of one pattern, and that the
1 Whilst dissenting from Dr Lawrence's results, I wish to express my keen
appreciation of his most useful essay. No one before Dr Lawrence seems to
have sufficiently appreciated the importance of Heinzel's work on the subject.
2 Pauls Grdr.fa) n, 1, 966-69.
Widsith and the Critics 141
same man could not be terse in writing a catalogue, diffuse in
singing a chieftain's praise ? If not, then arguments drawn
from dissimilarity of style cannot rank with arguments based
upon differences of tradition and outlook, such as we have seen
separate the catalogue of kings from the rest of the poem.
Excluding this catalogue of kings (18-49) and, of course,
the later interpolations, the poem seems homogeneous enough.
It is of course quite possible that Professor Brandl is right, and
that between our poem, and the mass of lays of the Eormanric
and other cycles which lie behind it, there may intervene some
heroic catalogues of champions or of peoples1. But I do not
think this has been demonstrated. It is not clear that the
same poet who depicted the figure of Widsith the traveller
could not have drawn up these lists for himself from the lays
which he knew.
However widely they may differ as to what are the old and
what the more modern parts of the poem, Professor Brandl and
Dr Lawrence agree in regarding the story of Widsith as ficti
tious. Mr Chadwick, in his recent study of the poem2, returns
to the old point of view of Guest and Haigh, Kemble and
Conybeare :
Any hypothesis which would represent the minstrel as a fictitious
character is open to the objection that, in that case, he would hardly have
been associated with so obscure a person as Eadgils, prince of the
Myrgiugas, a family not mentioned except in this poem3. On the whole,
then, the hypothesis that the kerneji of the poem is really the work of an
unknown fourth century minstrel, who did visit the court of Eormenric,
seems to involve fewer difficulties than any other. In that case, of
course, such passages as 11. 82 ff. must be regarded as merely the last
stage in a process of accretion which had been going on for some three
centuries.
1 Mnemonic verses similar to those of Widsith occur, as has been noted, in
Icelandic : and it is important to note that they occur in connection with the
same cycle of the struggle between Goths and Huns. It might plausibly be
argued that there existed in the late fifth or early sixth century a poem in which
this great fight was made the occasion of an "heroic muster roll" similar to
that which in later times was attached to the story of the battle of Bravalla :
that from this ancient poem much of the tradition of Widsith, of the Hervarar
Saga and of the ' ' Lay of Hloth and Angantyr " is derived, and that faint
echoes of it are to be found in the Elene and in Saxo.
2 In the Cambridge History of Literature, i, 36.
8 Whatever might be the force of this argument if any considerable propor
tion of O.E. heroic poetry were extant, it cn,n hardly count for much considering
how few fragments have survived.
142 Widsith
With this view I cannot agree. It is not merely that almost
every allusion which we can date — and there are many — shows
it to be a chronological impossibility that the poem, as we have
it, should record the real travels of a real scop1. This difficulty
might perhaps be evaded by a sufficiently thoroughgoing appli
cation of the interpolation-theory. But the visit to Ermanaric,
in which Mr Chadwick himself finds the original kernel of
Widsith's wanderings, is depicted in an atmosphere not of
history, but of heroic legend.
The Ermanaric of history ruled a mighty empire : only at
the very end of his long reign did the Huns appear, and he
slew himself before the contest between his people and the
Huns reached its climax. But in heroic story it is the chief
business of Ermanaric and his men to fight against the hosts of
Attila, who as a matter of history is younger by two genera
tions. And it is this heroic fiction which we see reflected in
Widsith. A real visitor to the court of Ermanaric would
probably have had nothing to say about the Huns, since their
ravages belong only to the last episode in Ermanaric's life.
Had he mentioned them, he could not have represented the
Goths as fighting them in the Vistula wood, which they had
left some generations earlier2 ; neither would he have repre
sented the Huns as " the people of Attila." Neither could he
have met in his visit to the land of the Goths the champions
whom he names. Even when we have condemned as inter
polations and excluded the names of ^Egelmund and Eadwine,
or even Theodric or Freotheric, our list is still an impossible
one. It includes Eastgota and his son Unwen, who had been
dead for generations when the Huns appeared ; it includes
Wudga, probably the foe, not of the Huns, but of the Sarmatians.
The mention of the Harlung brethren points to the legendary
Ermanaric, the slayer of his kin, rather than to the conqueror-
tyrant of history. The most important person in the poem is
Ealhhild, and if she be Eormanric's murdered wife, she belongs
to legend and not to history. Reject all these things as inter
polations, and what have we left ?
Again, the evidence that the catalogue of kings is older
1 See above, p. 5. 2 See below, p. 163.
Widsith and the Critics 143
than the story of Widsith's wanderings seems conclusive. Yet
the catalogue of kings can hardly be older than the sixth
century, since it includes the name of the Frankish Theodoric ;
to say nothing of other chiefs, like Attila, who are much later
than the period of Ermanaric.
Most of these facts have long been known : they were so
well known to Ten Brink and Miillenhoff that these critics
often assumed them to be known to their readers also, and
thought it unnecessary to instance them : a reticence which
has given an undue appearance of dogmatism to their state
ments.
These difficulties seem to me insuperable. Yet they are
only one portion of those that have to be faced, if we are to
believe that the kernel of our poem actually dates back to the
fourth century. But Mr Chad wick regards this date as in
volving fewer difficulties than the objection that, if the minstrel
were a fictitious character, he would not have been associated
with so obscure a person as Eadgils. Yet he has himself shown
how baseless this objection is. For if, as is exceedingly probable,
and as Mr Chad wick himself holds1, Eadgils is the Athislus of
Saxo's Fourth Book, then he can have been by no means an
obscure person, to have been remembered, eight centuries after
his death, in popular tradition, as a king " of notable fame and
vigour": vir fama studioque conspicuus*.
During the closing weeks of last year a study of Widsith
was published by Professor Siebs3. Whilst Chadwick regards
the Eormanric passages as essential, and the reference to
^Elfwine as later accretion, Siebs sees the kernel of the poem
in the lines on ^Elfwine, and regards the four passages in which
Eormanric is referred to as four separate additions.
Siebs strikes out as an interpolation 11. 88, 89 :
And ic wees mid Eormanriee ealle l>rage,
pser me Gotena cyning gode dohte,
together with the preceding passage. The name of ^Elfwine is,
1 See above, pp. 92-94, and Chadwick (135).
2 Saxo, iv (ed. Holder, p. 107).
3 Festschrift Vietor, Marburg, 1910, pp. 296-309.
144 Widsith
by this means, brought into immediate contact with the episode
describing the gift of the beag1. This, according to Siebs, is
the nucleus of Widsith : a short poem telling how (not Eormanric,
who has thus been cancelled, but) ^Elfwine gave the minstrel
a gift, how Widsith gave it on his return home to his lord
Eadgils, and how Ealhhild, sister of ^Elfwine, gave him
second gift.
What I do not understand is why such a poem as Siebs
constructs, in praise of ^Elfwine and his sister, should have
been afterwards interpolated and altered in the way he assumes.
I can understand an interpolator being supposed to add
additional episodes to an epic, additional names to a list, in
order to make them, as he conceives, more complete. But I
do not understand why, finding a poem in praise of the
generosity of JElfwine, whose glory was, as Siebs rightly urges,
so widely known, our interpolator should have deliberately
altered it, so as to make it celebrate the generosity of Eor
manric, who, though always reputed munificent, was certainly
also regarded in England, at this date, as a treacherous tyrant :
wra}> wcerloga, grim cyning*.
It has been urged by one of the greatest of Homeric critics
that, though a scholar may suspect certain passages to be
interpolated, he can hardly expect to convince others unless
he can show a reason why they were interpolated3. Siebs has
still to show why an ^Elfwine lay such as he postulates should
have been altered into an Eormanric lay.
1 In this Siebs follows Moller.
2 Siebs suggests that 11. 88, 89 may have been added to make a transition to
the innweorud Earmanrices (1. 111). But 11. 110-130, on his own theory, were
themselves added after 11. 88, 89. For they were, he thinks, added later than
the Introduction (1-9) (p. 807) ; and the Introduction is, he thinks, later than
11. 88, 89, and is based upon a misunderstanding of them (pp. 298, 300-301).
How then can 11. 88, 89 have been added to make a transition to the innweorud
Earmanrices (1. Ill, e£c.)?
Siebs' theory entirely depends upon his being able to prove 11. 88, 89 spurious.
Yet in order to account for the Introduction, he is compelled to admit them to
have been present in the poem at a very early stage. This admission naturally
renders any proof of spuriousness exceedingly difficult.
3 " Denn die Annahme einer Interpolation kann erst dann als erwiesen
betrachtet werden, wenn eine Veranlassung, die sie hervorrief, iiberzeugend
dargethan ist ; ohne diesen Nachweis bleibt sie ein subjectives Meinen, welches
vielleicht nicht widcrlegt werden, aber auch auf keine Beachtung Anspruch
machen kann." Kirchoff, Composition der Odyssee, 1869, p. 77. With this
Ten Brink, Beowulf, p. 3, may well be compared.
Widsitk and the Critics 145
And the objection that a passage cannot be dismissed by
simply declaring it to be an interpolation might also be raised
against Siebs' treatment of the Introduction (11. 1-9). Most
scholars will agree that Siebs is justified in regarding this as a
later addition : but they will hardly agree that he is therefore
justified in dismissing forthwith what we are told in the Intro
duction regarding Widsith's journey to Eormanric in the company
of Ealhhild. The Introduction seems to belong to the flourishing
period of Old English poetry, and its allusions are good evi
dence that, at the time it was added, Widsith's wanderings
were connected rather with Eormanric than with ^Elfwine1.
In face of these difficulties very heavy evidence should be
brought forward, if it is to be proved that all the Eormanric
passages are later additions to an original ^Elfwine lay. What
is needed is that a series of discrepancies between the Eormanric
and the ^Elfwine passages in syntax or metre or dialect or
heroic legend or geography, should be demonstrated, similar
to those upon which Mullenhoff and Ten Brink founded their
theories. But no such evidence has yet been produced.
The process postulated by Siebs has, then, no likelihood on
a priori grounds, and is supported by no positive evidence
Finally, it leads to rather incredible results. For, Siebs argues,
if the starting point of the poem was a song in praise of Alboin,
this gives a date too late to allow of a continental outlook :
Eadgils and his Myrgings must, he concludes, have dwelt in
England : they are perhaps identical with the Mercians, or at
any rate with the Mercian royal house.
Now there is a good deal of evidence which tends to place
the Myrgings on the continent2, but Siebs rejects this, on the
ground that some of the proposed identifications are not philo-
logically exact. Yet, in the case of remote names which are
based upon epic tradition, some degree of corruption may fairly
be expected, and page upon page of precedent for such corruption
might be, in fact has been, collected8.
But I do not understand how, having rejected identifications
1 See below, p. 151.
2 See above, Eadgils (pp. 92-94), and below, the Myrgings (p. 159) ; and cf.
notes on Fifeldor (1. 43) and Swcefe (1. 44).
3 See above, p. 23 : below, p. 160 : and cf. Heusler in Z.f.d.A. LII, 97.
c. 10
146 Widsith
such as those of the Myrging land with Maurungani, and Fifeldor
with the Eider, where corruption is easily intelligible, Professor
Siebs can approve, even tentatively, of an identification of the
Myrgings with the Mercians. For this is just one of those
cases where a phonetic discrepancy is difficult to account for.
The word Mierce must have been familiar to every eighth-
century Englishman. Why should this familiar word be altered—"
to the unintelligible Mierging, Myrging ?
But let this pass. To balance the abundant external
evidence favouring a continental home for the Myrgings, Siebs
produces none whatsoever tending to support the explanation
" Mercians." But on the other hand there is heavy evidence
against it. We are told " Meaca ruled the Myrgings " (1. 23).
Now we possess the genealogy of the Mercian royal house back
to Woden, but there is no Meaca there. And what of Eadgils
himself? This Mercian genealogy goes back to the tenth
generation before Penda. Is it likely that there could have
been, as Siebs thinks possible, an Eadgils, king of Mercia,
only a decade or so before the birth of Penda, that his fame
could have lasted in poetry into the eighth century or later,
and yet no trace of him remain either in the genealogies or
chronicle-entries ? And how can the Myrgings, if they are a
people dwelling in England, be spoken of either as identical
with, or neighbours of, the Swoefe (i.e. Suevi) ? And how can
Offa, king of Angel, ancestor of the Mercian house (in whose
figure in Widsith Prof. Siebs thinks there may even be
reminiscences of the historic Offa II, king of Mercia), be said
to have marked off the borders of his kingdom wi\ Myrgingum,
if this means ' against the Mercians ' — his own people ?
Recent scholarship has been rather too ready to dismiss the
conclusions of earlier students without sufficiently examining
the facts from which those conclusions were drawn. In this, as
in some other things, it hardly compares favourably with the
school of Mtillenhoff, Moller and Ten Brink.
For each of these critics based his work upon a careful
study of his predecessors' investigations, and a general accept
ance of their results. Ten Brink's theory represents in its main
Widsith and the Critics 147
outline the gradual evolution of opinion : only its details are
the hazardous guesses of a single scholar. The view which
I have expressed above as to the structure of Widsith is little
more than an adaptation of Ten Brink's in its main outline,
allowing for the discoveries which have been made since his
time, and for certain publications issued shortly before he
wrote but which he appears to have overlooked. I regret that
for this opinion I cannot claim the support of many recent
critics, to whom every student of Old English is so much
indebted : but I have been pleased to find that it is the
view of Dr Axel Olrik. The poem has, Dr Olrik thinks, with
reason been regarded as the oldest poem of the German race,
yet more rightly it may be regarded as a concretion of several
pieces of old poetry : the first of these is the " Catalogue of
Kings " to which the lines on Offa and the lines on Hrothgar
were subsequently added1.
Indications of Date.
In endeavouring to date Widsith, two extreme courses are t/
open to the critic. He may refuse to take into consideration
the possibility of interpolation, and so insist upon judging the
poem exactly as he finds it. In this case he will clearly have
to date it somewhat late.
This, we have seen, was the view of K. Maurer :
•I cannot persuade myself that a poem which collects in the baldest way
the names of heroes and people, and withal goes so far as to mingle in the
most motley style with the figures of Germanic history and saga Israel
ites and Syrians, Hebrews, Indians and Egyptians, Medes, Persians and
Idumaeans, Saracens and Seres, Alexander the Great, the Emperor of the
East, and a second emperor... can really date, as Miillenhoff assumes, from
the seventh century2.
Maurer accordingly suggested the age of the Vikings or the
beginning of the ninth century as more in accordance with
the phenomena of the poem. A similar view had previously
been taken by E. Jessen3, who dismissed the poem as a mere
1 Helte-Digtning , 11. I cannot say whether Dr Olrik would agree with my
other contention that it is not possible to divide Widsith further into component
lays, but only to mark off certain later interpolations.
2 Z.f.d.Ph. n, 447 (1870).
Undersflgelser til nordisk oldhistorie, Ktfbenhavn, 1862, p. 51.
10-2
148 Widsith
promiscuous collection of some monk or priest, and very similar
was the opinion subsequently expressed by F. Ronning1 and by
A. Erdmann2.
But none of these students had devoted any special atten
tion to the poem. They are concerned with maintaining
theories of different kinds against which Widsith might be
quoted as evidence, and they proceed accordingly to disqualify
that evidence by judging Widsith on the merits of its most
doubtful passage. It is a case of abusing the plaintiff's attorney.
Naturally, a scholar who is seeking to rewrite history, and
in spite of Tacitus, Bede, Nennius and king Alfred, to make
the Angles an inland tribe, finds Widsith one of the first
impediments which he must clear out of his way3. Our confi
dence in Widsith should be strengthened when we find that,
the more paradoxical a critic's views are, the more anxious he
is to attribute a late date to the poem.
But if some critics have refused to allow for interpolations,
others have surely pressed the theory of interpolations too far.
They have started with the assumption that the poem, or at
any rate the original kernel of it, is the work of a contemporary
of Ermanaric, and have then dismissed as interpolated every
passage which would not harmonize with this date. This, as
we have seen, is the view of most English critics : of Conybeare,
Kemble, Guest, Haigh, Stopford Brooke, Earle, Garnett and
Chad wick4. The objection to this theory is that it compels
us to dismiss as interpolations almost all the historical data
(and they are many) which the poem affords6.
The extreme dates attributed to Widsith are, then, nearly
* five hundred years apart. Yet it would seem possible to arrive
at some approximately accurate results, upon which the majority
of critics are agreed. " A mean there is between these extremes,
if so be that we can find it out."
1 Beovulfs-Kvadet : En literar-historisk undersfgelse, Kefbenhavn, 1883.
WidstiS...er sikkert et temmelig sent forfattet "register" (p. 3; note 2): for-
fattet vi ved ikke n&r (p. 104).
2 Heimat der Angeln, p. 50.
3 See Appendix, Note E, On the original homes of the Angli and Varini.
4 See above, p. 4, and cf. Lawrence in Mod. Phil, iv, 355-6.
8 See above, p. 142.
Widsith and the Critics 149
Some criterion is wanted by which we can judge of inter
polations. Perhaps we may accept the principle that, in
investigating the date of the poem, we have no right to reject a
passage as interpolated on the sole ground that it does not
agree with our view of the date ; for to do this is to argue in a
circle. On this principle, we ought not to dismiss a passage
like that relating to ^Elfwine, if in spirit, metre, and style it
is exactly in harmony with the rest of the poem ; but we should
be quite justified in refusing to admit any argument as to date
which is based on 11. 75-88. These lines are full of Biblical
names, whilst elsewhere we meet with no Biblical names
whatever in the poem. Some of them cannot be scanned,
whilst elsewhere the poem is metrically accurate. The whole
spirit and atmosphere is different. We must then either reject
the lines altogether, or isolate them as suspect.
Every reader must be struck by the knowledge of conti
nental conditions displayed by the poet ; he knows the most
ancient figures of Gothic tradition, Eastgota and Unwen, and,
to judge from the way in which he introduces their names, he
expects his audience to be intimate with their story. Traits
such as the coupling together of Rugians and Romans seem
reminiscent of the early sixth century. The mention of the
Goths fighting in the Vistula wood is a reminiscence of an
earlier period still1. The poet's interest seems to centre in //
continental Angel.
Hence so many critics have argued that the poem was!
composed before the conquest of Britain. This however by no!
means follows. Even if Widsith's visit to the Gothic court does'
witness to a time when that court set the fashion in minstrelsy
to Western Europe (and we have seen that the evidence for
this Gothic preeminence in song is insufficient2) this would not
prove that Widsith must have been composed at that time3.
In a Mercian hall, in the opening years of the seventh $?
century, some of the older men would have been of continental
1 See also notes to 11. 121-122.
2 See above, p. 13.
8 Bojunga (P.B.B. xvi, 548) puts die dlteren bestandteile des WidstiS in
a time when the Gothic court was still preeminent.
150 Widsith
birth, and many more would be the sons of those who had been
born in Angel. The songs sung in the hall would still be the
old continental lays, the traditions would be those of the sixth
century or earlier, of Goth and Hun. This is all that the data
of Widsith demand.
Putting aside the criticisms which would date the kernel
of our poem in the fourth century, and those which will not
allow it to antedate the ninth, we find a considerable measure
of agreement on the part of the more moderate critics. Widsith
is the oldest English poem (Miillenhoff1, Gum mere2), which is
equivalent to saying the oldest poem of any Germanic people
(Olrik8) ; it reflects the traditions of the time of the migrations
(Jiriczek4) ; traditions which the Angles must have brought
with them to England (Leo5, Wiilcker6, Panzer7) ; belongs in its
I/essential part mainly to the seventh century (Symons8, Mb'ller9,
Miillenhoff10, Ten Brink u) ; incorporating in the catalogue of
kings an earlier poem belonging to the close of the period of the
migrations, the sixth century (Olrik12, Moller13, Miillenhoff14,
Ten Brink16).
These students have based their results upon a study of the
heroic traditions of the poem, including geography only in so
far as that is necessarily involved by the heroic traditions.
We may, then, perhaps outline these as the results which
follow from a study of the heroic lore of Widsith :
We have an exceedingly early poem, belonging probably to
the seventh century, but reflecting the traditions of the fifth
I Z.f.d.A. xi, 294. 2 M.L.N. iv, 418.
3 den gotiskefolkestamm.es celdste digtning. Helte-digtning, 11.
4 Jiriczek d) 110 ; (<>) 101. 6 Sprachproben, 76.
6 Grundriss, 329. 7 H.G. 436.
8 Pauls Grdr. (2) in, 621.
9 V.E. 18, where Moller puts " das eigentliche Widsith-lied " after A.D. 580.
10 Beovulf, 93.
II Pauls Grdr. (i) n. Ten Brink puts the contamination of the Eormanric-
Ealhhild lay, which constitutes the main poem, in the late seventh century (544),
though he thinks the constituent parts earlier (541).
12 forfattet omkring folkevandringstidens slutning. Helte-digtning, 11.
13 V.E. 18 : aus der mitte oder dem anfang des sechsten jahrhunderts.
14 Beovulf, 92, 93. 1S Pauls Grdr. d) n, 1, 538.
Widsith and the Critics 151
and sixth, and incorporating one piece of verse, the Catalogue
of Kings, which seems older than Widsith proper.
Widsith has been interpolated : and though it would be too
ambitious to hope that we can define exactly the limits of the
interpolations, it may be well to make a list of the passages
open to suspicion, upon which argument should therefore not
be based, without remembering that they are probably later
accretion. In this schedule of suspected passages we must
place the Biblical interpolation, with the other lines condemned
by Miillenhoff (11. 14-17, 75-87, 131-134)1, and the not dis
similar lines which occur in the list of the champions of
Eormanric2 (11. 117-118).
In view of the way in which the other secular poems of the
Exeter book have been " edited," and of the very similar pro
logue and epilogue with which the Wanderer is furnished, we
are bound to regard the prologue and epilogue of Widsith with
some suspicion3. But I do not think we are therefore justified
in dismissing as worthless what we are there told. Granting
the prologue to be a later addition, its metre and style show
it to belong to the best period of old English versification.
It tells us things about the Traveller which could not have
been gathered merely from the extant poem as we have it.
It seems reasonable to conclude that, even if the Prologue is
a later accretion, its author was using traditions of which we
now have no other record4.
The lines on ^Elfwine form a problem which has met us
more than once. In date there is a considerable period between
Alboin and the other heroes of the poem6, but, as pointed out
above6, we cannot put these lines aside on this score alone. If
a more minute examination of grammar and metre shows this
passage to have peculiarities which are also common to the
other suspected lines, but which are not shared by the un
suspected passages, then we must regard it also as interpolated.
1 See above, p. 8.
8 See above, p. 122. These lines were rejected by Moller.
8 So Moller, etc.
4 See above, pp. 28, 145.
6 Miillenhoff (Nordalb. Stud, i, 149) suggested that this passage was perhaps
an interpolation.
• p. 149.
152 Widsith
Otherwise we must regard it as original : meantime it is
doubtful.
We have then:
The Catalogue of Kings : 17 11. A
With lines on Offa, Hrothgar, Hrothwulf and I ^ ,.
Ingeld: 15 11.
Widsith proper : 65 11.
Possible later interpolations : 46 11.
These results as to date and interpolations are, of course,
provisional, and must be checked by any other information
which we can procure.
In the first place we may be able to control them by a
systematic survey of the geography of the poem. This will
involve a certain amount of repetition, as we have already had
to consider geographical points necessarily involved in the
study of heroic legend. But it is worth while to see what
results we can draw from a comparison of all the geographical
data of Widsith.
Secondly, we may be able to check these results by a study
of the grammar, dialect and metre. Here especially we may
hope for some additional light, for in these matters more
progress has been made since the days of Mulleuhoff than in
the investigation of the ancient geography of the coast tribes.
CHAPTER V.
THE GEOGRAPHY OF WIDSITH.
HOWEVER wide his wanderings, no poet could know equally
well the national sagas of all the Germanic tribes, and we shall
find that the stories known to Widsith are those belonging
particularly to two of the five great divisions of the Germanic
race, as it existed in the early centuries of our era.
But in any attempt to classify, into their groups and sub
groups, the tribes and heroes of Widsith, we are met at the
outset by the difficulty of an unsettled terminology. The
result is that this chapter cannot be written as simply as
might have been wished.
Linguistic Classifications.
In trying to realize these ethnographical divisions of ancient
Germany, scholars have had to rely chiefly upon linguistic
evidence, and this evidence only gives up its secrets very
gradually. More and more accurate results are obtained, but
even the final result is only an approximation to the truth.
Hence the old classification — Scandinavian, Low German,
High German — has been superseded by that of Scandinavian
or North Germanic, West Germanic (embracing both Low and
High German) and East Germanic (embracing Gothic and the
other kindred tongues).
Yet even this perhaps does not represent ultimate truth.
The so-called " Northern " and " Eastern " families are possibly
more intimately interrelated. Even if the purely linguistic
evidence is not strong enough to establish this connection —
and many scholars have held that it is — there remains the
154 Widsith
significant fact that the same tribal names, Rugian and
Vandal, East Goth and West Goth, occur in both the
Northern and the Eastern family, but not in the West. One
of the minor difficulties of Widsith is to decide when the
reference is to the " Northern " and when to the " Eastern "
people of the same name.
Nor must we forget that the apparent homogeneity of the
West Germanic group may be deceptive. It has been argued
that the West Germanic peculiarities are a development of the
early centuries of the Christian era. Barred by the Roman
legions from spreading, as their fellows in Northern and Eastern
Germany might do, a number of distinct German tribes were
pressed together between the Elbe and the Rhine. During
these years, it is urged, peculiarities were developed which
have overlaid and concealed earlier dialectal differences. We
group these tribes together as West Germanic, but it may well
be that the difference between the so-called Anglo-Frisian stock
of the North-Sea coast and the inland tribes is ancient and
deep-lying1.
In dealing with Widsith it is of particular importance to
keep in mind this distinction between the "Anglo-Frisian"
tribes of the coast, and the inland Deutsch, or Germans proper.
But, owing to peculiarities developed in the High German
dialects of the South, this Deutsch or German speech was itself
soon cloven into two great sections, the so-called High and
Low German. Now English students have generally, in prac
tice, applied the term "Low German" to all those West
Germanic dialects which did not undergo the High German
sound changes : whilst the Germans have kept in mind that,
obvious and striking as are the distinctions which differentiate
the High German from the other West Germanic dialects, yet
those more obscure differences which separate the Anglo-Frisian
from the German proper are more ancient and fundamental.
" Low German " therefore, in the terminology of English
scholars, generally includes English and Frisian, whilst Nieder-
deutsch in the terminology of German scholars excludes them.
1 See Bremer in Pauls Ordr. ^ HI, 809 ; Bremer in I.F., iv, 8-31 ; Siebs in
Pauls Grdr.M i, 1154.
The Geography of Widsith 155
I propose to use " Low German " as excluding the Anglo-Frisian
dialects. For to do otherwise is to overlook the early difference
between the coast tribes and the Germans of the inland plains ;
a difference which is demonstrable philological ly, and which is
important to the student of the geography and heroic tradition
of Widsith.
We have then East Germanic, North Germanic, and three
divisions of West Germanic : Anglo-Frisian, Low German and
High German.
The Roman classification, and its relation to the linguistic
divisions.
To harmonize the linguistic classification with that of the
Roman writers is not easy. Tacitus' famous threefold division
of the Germans cannot be reconciled with modern philological
classifications unless we suppose that, knowing little of either
the North or the East, he left these regions out of his con
sideration, and intended his1 classification of the Germans into
Ingaevones, Herminones and Istaevones to point only to the
three great sections of the Western folk. On this theory the
Ingaevones2 [Ingyaeones3] who dwell proximi Oceano are what
philologists call the "Anglo-Frisians": the North-Sea folk.
The Istaevones2 [Istraeones3] are defined by Pliny as proximi
Rheno, they are therefore the predecessors of the group of
peoples later known as the Franks. The Herminones2
[Hermioues3] stretch further inland : they are medii2 or
mediterranei3, and so coincide with that great stock which
subsequently, spreading further south, became the High
German people. That Tacitus should have left the Eastern
and Northern tribes out of account seems strange. Pliny
mentions the East Germanic group by name. Owing to the
prominent part which the Gothic tribe has played in history,
and to the accident that their language, alone of the East
Germanic tongues, has been preserved, we think of the Goths
1 The classification was, of course, derived by Tacitus from the tribes of the
lower Rhine, and represents their point of view. Cf. Olrik in Folk-Lore,
xix, 398.
2 Germania, n. * Pliny, Nat. Hist, iv, 14 (28), ed. C. Mayhoff.
156 Widsith
as the typical East Germanic people1. But Pliny applies to
the Eastern group the name Vandili2: a name which subse
quently came to be restricted to one particular tribe of the
group, the conquerors of Africa3. For the North Germans no
satisfactory collective name has ever been found. They are the
dwellers in the " island " of " Scandia " or " Thule," and classical
writers pass, hastily and with a shudder, to more congenial
topics.
This classification may be expressed in tabular form. Bu
a table is too dogmatic: it compels us to draw hard and fast
lines between groups of dialects when these dialects, in point
of fact, probably shaded imperceptibly into one another. When
we speak of Anglo- Frisian we must not imagine one homo
geneous speech with definite boundaries, but a number of
dialects grouped together in virtue of common characteristics.
And these dialects had also other connections : the Anglian
speech had affinities with the neighbouring Scandinavian, the
Frisian is stated to have had similar affinities with the neigh
bouring Low German4.
1 Already in the sixth century the general name of the " Gothic races" has
come to be applied to the East Germans: ForBiKa tdvij iro\\a (itv Kal a\\a
irpbrephv re r/v Ka.1 ravuv ecrn, ra 8£ di] iravruv fjityurTa re Kal a^ioKoydn-ara. TbrQoi
T£ elffi Kal f$avdi\oi Kal OvuriyorOoi Kal T-/iTrai5es...<j><i)vri re atrois &TTI pia, TorBtKi)
\eyo/j.frri, Kal /xoi 5oKovt> tt- Ms fj.ev elvai airavres rb ira\<ubv ZQvovs. Procopius,
Bell. Vand. i, 2.
2 Tacitus refers to this group-name without explaining it.
3 The interpretation of Tacitus' nomenclature in the terms of modern
philology is, of course, highly conjectural. But it has the advantage of
enabling us to attach a meaning to Tacitus' classification, which otherwise
would be unintelligible to us, and it is convenient. Hence it has been widely,
and sometimes too dogmatically, accepted.
On the other hand, a recent English critic is perhaps unduly sceptical.
"The whole scheme indeed seems to me to be based on a fundamental error.
The sound-changes which differentiated the Scandinavian, Anglo-Frisian, and
German groups of languages from one another appear to have operated in the
fifth, sixth, and seventh centuries. On the other hand the ethnic groups called
Istaeuones, Inguaeones and Heriniones, seem to have been obscure and probably
antiquated in the first century." (Chadwick, 223.) Now this criticism would
be quite just if levelled against any attempt to equate the classification of
Tacitus with that of the philologists of half a century ago. Their classification
often did rest upon changes of the fifth, sixth and seventh centuries. But the
classification now in use is founded upon those much older linguistic differences
which have only gradually been revealed, because they were overlaid and con
cealed by the drastic sound-changes of the later centuries. It may well be that
some of these older linguistic differences go back to the time to which the
classification of Tacitus refers. See Bremer in I.F. , iv, 8-31.
4 See Morsbach, Beiblatt to Anglia, vn, 323-332 ; Siebs in Pauh Grdr.^ i,
1154-7 ; Bremer in I.F. iv, 8, etc.
The Geography of Widsith
157
With this caution we may classify the Germanic dialects
thus :
I. EAST GERMANIC. Latin and Greek writers use Gothic or Vandalic
to cover the whole of this group.
II. NORTH GERMANIC or SCANDINAVIAN.
A. ANGLO-FRISIAN (perhaps corre
sponding to an earlier Ingaevonian
These two classes
group of tribes).
being free from
III. WEST
GERMANIC. '
B. DEUTSCH
f Low GERMAN proper
(NiederdeutscK), per
haps corresponding
to earlier Istaevonian
the High German
• peculiarities are
often inaccurately
grouped together
as " Low German.''
or GERMAN
(Prankish) group. '
in narrower
sense.
HIGH GERMAN (Hoch-
*
deutsck), perhaps cor
responding to earlier
k. Herminonian group. .,
Later Movements.
The terms ' East ' and ' West ' have reference, of course, to
the original positions of these groups. By the sixth century
the Eastern people had migrated, and were South and West of
the Western people. One of the extraordinary things about
Widsith is that its author's geographical lore seems often to
date back to a period before these migrations.
The East Germans — Goths, Gepidae, Burgundians and
Vandals — battered down the Roman Empire and divided the
fragments among themselves ; but they perished in the process,
and their widely scattered tribes were soon absorbed in the
Romanized populations among whom they settled ; they were
defeated piecemeal by later conquerors, and have left no modern
representatives. A Gothic remnant maintained themselves
throughout the Middle Ages on the mountainous southern
coast of the Crimea — a little nation of a few thousand souls,
subject to a Tartar prince. At the end of the sixteenth century
they disappear from history1, and with them the East Germanic
races end. The North Germans, on the other hand, in spite of
their plundering viking raids at a later period, have mostly
1 Gislenii Busbeqaii, Epistola, Parisiis, 1589, pp. 135-7. Cf. Life and
Letters of Ghiselin de Busbecq by Forster and Daniell, 1881, pp. 355-9.
158 Widsith
remained in Gothland, Sweden and Norway — their ancient
seats. Of the West Germans, the sea tribes, who once held
the coast from the Zuider Zee to the Baltic, have paid the
penalty of their piracy. Angles and Jutes conquered Britain,
and Danes occupied their ancient continental homes. The
Lombards — if indeed they are to be classed with this group —
left the lower Elbe, and were lost amid the Italians, while the
Saxons, besides helping in the settlement of England, first
conquered and then absorbed various inland tribes dwelling
in the modern Hanover and Westphalia. The continental or
" Old " Saxons therefore ceased to be purely Anglo-Frisian in
blood or speech, and have rather to be counted as "Low
German." So that the "Anglo-Frisian" speech is now repre
sented only by a few hundred thousand Frisian-speaking folk
in the north of Holland and in the islands of the North Sea,
and by English-speaking people in other parts of the world.
The Istaevones, if we equate them with the Franks, became
the ancestors of the Dutch, of the Flemings and of the
Germans of the Lower Rhine : these " Low Frankish " together
with the " Low Saxon " dialects constitute the " Low German "
proper. The Herminones of Roman times included the Her-
munduri, Marcomanni and Suevi : great nations which, with
some changes both of boundaries and name, are represented by
High German peoples — Thuringians, Bavarians, Swabians.
But though it may be possible to trace some measure of
continuity in the different classifications of the Germanic
people from the days of Tacitus onwards, it is better not to
follow certain German scholars in speaking of an Ingaevonian
dialect when we mean Anglo-Frisian. That the two terms
corresponded is, after all, but a theory; that they corresponded
exactly is unlikely.
Although the many peculiarities common to Old English
and Frisian point to intimate intercourse between Angles and
Frisians in the early centuries of our era, this does not in the
least preclude intercourse, on the other side, between Angles
and Danes. This communication has left its mark in certain
phonetic peculiarities which Anglian possesses in common with
the Scandinavian tongues. Anglian bards, we have seen, were
The Geography of Widsith 159
interested in the fight at the Danish Heorot, as well as in the
fight at the Frisian Finnesburg. The Baltic, like the North
Sea, would unite rather than separate seafaring nations.
It is quite likely, then, that the Ingaevonians may have
included not only " Anglo- Frisian " but also Scandinavian
tribes1 : and, however this may be, the intercommunication
between the most northern of the Anglo-Frisians and the most
southern of the Scandinavians is obvious. Evidence of it meets
us everywhere in Widsith.
The Traveller and his people — The Myrgings.
More than once in our poem mention is made of the
Myrgings, a tribe to which, it would seem, Widsith himself
belongs. Unfortunately the text is not clear. Hine from
Myrgingum celpele onwocon cannot be construed ; but, what
ever be the correct reading, some connection between the scdp
and the Myrgings is indicated. In a latter passage, too, Widsith
speaks of the prince of the Myrgings, Eadgils, as his liege-lord.
Who these Myrgings were it is not easy to say. We
should gather that, on the north, their boundaries touched
those of the Angles, for a frontier dispute was settled by Offa
with the sword at Fifeldor — i.e. the river Eider, which is still
the boundary between Schleswig and Holstein2. Again the
word Swcefe seems to be used as synonymous with Myrgingas.
We should conclude then, from the poem, that the Myrgings
dwelt south of the Eider, and that they were members of the
wide-spread Suevic stock.
From other sources little help is to be got. A Mauringa is
mentioned by Paul the Deacon3 as having been visited by the
Lombards early in their travels, whilst the " Geographer of
Ravenna4," who is supposed to follow an original composed
1 For the evidence see Schiitte, A.f.d.A. xxvm, 9-10; Kossinna in I.F.
vii, 308-11; Chadwick, 209, etc., 295-6.
2 See note to 1. 43.
8 Igitur Langobardi tandem in Mauringam pervenientes...egressi itaque
Langobardi de Mauringa applicuerunt in Golanda. Paulus i, 13 in Script. Rer.
Lang. ed. Waitz, in M.G.H., 1878.
4 Ravennatis anonymi cosmographia, ed. Finder andParthey, Berolini, 1860,
pp. 27, 28 (§ i, 11).
160 Widsith
in the sixth century, twice mentions Maurungani: gani is,
I suppose, miswritten for gaui (cf. Goth, gawi, " country ").
The attempt of MiillenhofF1 to make the Myrging name
coincide linguistically with Mauringa, Maurungani was hardly
successful: we should expect an O.E. form Miering, not
Mierging. In the most elaborate recent study of early German
geography the identification has therefore been abandoned
altogether2, whilst those who have adhered to it have had to
admit that the names do not linguistically quite correspond,
and that corruption of some kind has crept in3.
It is perhaps permissible to point out that g, especially in
the neighbourhood of a u, is a difficult sound to be certain
about, and that a somewhat similar intrusive g, which at one
time seemed as inexplicable as the g of Myrging, has since
been shown to be in accordance with phonetic laws4. But let
it be granted that the g of Myrging is " uuphonetic." It does
not therefore follow that the identification of the two words is
"impossible on philological grounds5." It is true that in
dealing with the native words of a language, the presumption
is that such words will only change in accordance with phonetic
law. But this does not apply to legendary and foreign
geographical names which, as they pass from mouth to mouth,
often undergo transformations which are the result, not of
phonetic law, but of mere error6. Are we to account phonetically
for every letter in such a change as Stafford for Oxford, through
an intermediate Asquesufforch, in Berners' Froissart ?
1 Z.f.d.A. xi, 279-80. See also Moller V.E. 28, 29 and A.f.d.A. xxn, 152.
Holier would connect the Myrgings with the 'M.apovivyot of Ptolemy [n, 11, 22]
and the Marsigni [for *Marvigni] of Tacitus.
2 Much in P.B.B. xvn, 192-4. Attempts to account for the name etymo-
logically will be found there, and in A.f.d.A. xvi, 24 (by Kossinna). Cf. Hirt
in P.B.B. xxi, 148.
3 Thus Heinzel suggests that the g may be due to a popular etymology con
necting the name with mirige "merry"; see Ostgotische Heldensage, in W.S.B.
cxix, 3, p. 25, 1889.
4 By Sophus Bugge in P.B.B. xni, 504 " Germanisch ug aus uw." The g or
ug seems generally, though not always, to have developed out of an earlier
uw. Cf. O.E. geoguft: Gothic junda; O.E. sygel: Goth, sauil; O.E. nigon:
O.H.G. niwan; O.E. sugu: O.E. su. See A. Noreen, Abriss der urgermanischen
Lautlehre, Strassburg, 1894; Kluge, Vorgeschichte der altgerm. Dialekte, in
Pauls Grdr.w i, 380; Moller in A.f.d.A. xxii, 152 (footnote).
8 Lawrence in Mod. Philol. iv, 362, following Heinzel, Ostgot. Heldensage,
25, and followed by Holthansen, n, 166.
6 See Heusler in Z.f.d.A. LII, 97, etc. Heldennamen in mehrfacher Lautgestalt.
The Geography of Widsith 161
It does not appear that Paul himself was certain where
Mauringa was : in his account of the early Lombard wanderings
he blindly followed older authorities, and was obviously puzzled
by the names of tribes and places which he reproduced1. Of
the " Geographer's " two references, the one is vague, and the
other corrupt2. It seems clear, however, that this Maurungani
is a district bordering on the Elbe, and apparently extending
indefinitely in an eastward direction over those lands which
the Germanic tribes had vacated when they broke into the
Roman Empire3. Northward it borders on the Danish country
(Dania). It therefore includes Holstein, together with an
uncertain amount of what is now Eastern Germany. Here
too, east and north of the Elbe, it seems best to place Paul's
Mauringa ; and certainly it is here that we must place the
Myrgings, south of the Eider, in the modern Holstein.
Both phonetically and geographically Myrging sufficiently
coincides with Maurungani to make the identification highly
probable. But it would not be safe to base argument upon it.
Geographical classification of the names in Widsith.
An attempt to classify geographically the tribes, localities
and heroes mentioned in the undoubted portions of Widsith
gives this result.
EAST GERMANIC.
Tribes and Places : [HKETH-]GOTAN, BURGENDAS*, GEFTHAS, WISTLA-
WUDU.
Heroes : JEormanric, Becca, Gfifica, Guthhere*, Herelingas, (Emerca and
1 Most of the names come from the Origo gentis Langobardorum : Mauringa
however is not found there.
a See Appendix B, Maurungani and the Geographer of Ravenna.
8 Zeuss (472) and Schmidt (Zur Geschichte der Langobarden, 1885, p. 47)
identify Mauringa with [Maurungani] which they place east of the Elbe.
Bluhrne (Die Gens Langobardorum und ihre Herkunft, Bonn, 1868, p. 23)
placed Mauringa to the west, but on quite insufficient grounds. Mullenhoff,
who so read the poem that Eadgils, prince of the Myrgings, was the husband
of Ealhhild, made the Myrgings extend towards the old Lombard home in
Pannonia, in order to supply a motive for the match : which would, he thought,
have been unreasonable had the Myrgings not extended beyond Holstein. See
Z.f.d.A. xi, 279; Beovulf, 99; Nordalb. Stud, i, 140, etc.
4 Geographically, the Burgundians are, when we first meet them, indis
putably East Germanic. Evidence for the character of the Old Burgundian
speech is scanty, but is generally considered to confirm its East Germanic
0. 11
162 Widsith
Fridla), East-Gota, Unwen, Sifeca, Hlithe, Incgentheow, Wynnhere,
Rumstan, Gislhere, Freotheric, Wudga, Hama.
%* Probably we should include WULFINGAS : Helm ; also Theod-
ric, Seafola, Heatkoric.
The following are spoken of as Gothic champions, and may
therefore be classed here, though we have no information outside
our poem for so placing them : Hethca, Secca, Wulfhere, Rcedhere,
Rondhere, Withergield.
NORTH GERMANIC.
Tribes and Places : DENE, S.E-DENE, SUTH-DENE, SWEON, GEATAS,
WENLAS, HEATHOREAMAS, HEOEOT.
Heroes : Sigehere, Alewih, Hrothwulf, Hrothgar, Ongendtheow.
*#* Perhaps we should include THROWENDAS.
ANGLO-FRISIAN.
Tribes and Places : W^ERNAS, EOWAN, YTE, FRESNA CYNN (FRYSAN),
HOCINGAS, SYCGAN, ENGLE (ONGEL), SEAXE, FIFELDOR.
Heroes : Billing, Oswitie, Gefundf, Finn, Folcwalda, Hncef, Sceferth^ Of a,
Beadeca.
%* Probably we should include WOINGAS, WROSNAS : Wold,
Holen.
LOW GERMAN.
Tribes : FRONCAN, HETWARE.
Heroes : Theodric, Hun.
%* Probably we should include THYRINGAS : Wod.
HIGH GERMAN.
Tribes: ? ^ENENAS, ? THYRINGAS. (Both very unlikely.)
Heroes: ? Wod. (Very unlikely.)
The following tribes and heroes cannot be classified exactly,
but it is clear that they are all maritime, and are to be localized
somewhere on or near the Baltic :
Tribes : HOLMRYGE, GLOMMAS, H^ELSINGAS, YMBRE, LONGBEARDAN (1. 32),
WlCINGA CYNN, HEATHOBEARDAN.
Heroes : Hagena, Heoden, Wada, Sceaf there, Sceafa, Ingeld.
*#* Near the Baltic we must also place the BRONDINGAS,
HUNDINGAS, MYRGINGAS, SW^FE : Breoca, Mearchealf, Mearca,
Witta, Eadgils ; perhaps WULFINGAS : Helm ; the HEREFARAN and
Hringweald on either the North Sea or the Baltic.
character. Hempel (Linguistic and Ethnografic status of the Burgundiant :
Trans. Am. Philol. Assoc. xxxix, 1909, pp. 105-119) argues for Anglo-Frisian
and Norse affinities. But this theory rests upon the interpretation of certain
runic inscriptions the reading of which is exceedingly doubtful, and which in
any case cannot be proved to be Burgundian.
The Geography of Widsith 163
The non-Germanic people mentioned in the undoubted parts
of Widsith all belong to the Mark :
Tribes : HDNAS, CBEACAS, FINNAS, WINEDAS, RUMWALAS.
Heroes : ^Etla, Casere, Gaelic.
The following are quite uncertain :
Tribes : BANINGAS, RONDINGAS, GEFFLEGAS, SWEORDWERAS, HRONAS,
DEANAS, FRUMTINGAS.
Heroes : Thyle, Sdlling, Ealhhild daughter of Eadwine (unless the
reference to ^Effmne son of Eadwine be genuine, in which case it
is probable that Ealhhild is regarded as a Lombard princess, sister
of ^Elfwine).
The following names occur in the suspected passages only :
Tribes and Places : " EASTAN OF ONGLE," EATUL, SERCINGAS, SERINGAS,
WALA RICE, SCOTTAS, PEOHTAS, SCRIDE-FINNAS, LIDWICINGAS, LEONAS,
HiETHNAS, H^lRETHAS, ISRAHELAS, EXSYRINGA8, EBREAS, INDEAS,
EGTPTAS, MOIDAS, PERSAS, MOFDINGAS, ONGEND-MYRGINGAS (?),
WITH-MYRGINGAS (?), AMOTHINGAS, EAST-THYRINGAS, EOLAS, ISTE,
IDUMINGAS.
Heroes : Widsith, Hwala, Alexandreas, JElfwine son of Eadwine, Elsa,
jEgelmund, Hungar.
In these lists we notice that
(1) a wide knowledge is shown of Gothic story. But this
is legendary, as is all the knowledge our poet shows of East-
Germanic tribes : it sometimes reflects a state of things which had
passed away even in the fourth century. Thus the localization
in the Vistula-wood of the struggle between Goth and Hun is
unhistoric ; in Ermanaric's day the Goths were dwelling on the
plains near the Black Sea, and it was on these plains, between
the Sea of Azov and the Danube, that the struggle took place-
The term Hrcedas, applied to the Goths, is apparently legendary,
not historic1. The grouping of Burgundians and Goths together
is legendary geography2.
(2) Of the North Germanic tribes there is most knowledge
of those bordering on the Baltic, and especially of the Danes.
(3) Full information is shown of the Anglo-Frisian tribes
and heroes, and also
1 See Appendix H.
2 See note to 1. 19. Schiitte (Oldsagn, 46) finds a similar archaism in the
grouping of Gepidae and Wends, which he thinks points to " Gepidernes
gamle Bopaale ved 0sterstfen." But this is not conclusive : for the Wends
moved southward also, and still bordered on the Gepidae after these latter
had settled in Dacia. It is clear from Jordanes, v (33, 34), that the two tribes
were still immediate neighbours ia the sixth century.
11—2
164 Widsith
(4) of a large number of tribes which we cannot place
exactly, but which clearly dwelt upon the coast, in some cases
perhaps of the North Sea, but more generally of the Baltic.
(5) Of Low German tribes and heroes there is little
cognizance. Of the many Frankish heroes Theodoric (the son
of Clovis) alone is mentioned : and we happen to know that
Theodoric's victories were famed not only among his own
people, but also among the coast-tribes. The other Low
German peoples mentioned, the Hetware, and probably the
Thyringas, were immediate neighbours of these coast tribes.
(6) Of High German tribes and heroes there seems to be
no knowledge whatever. For the identification of the ^Enenas
with a Bavarian family is more than doubtful, and the Thyringas
are apparently the Low German Thoringi. The absence of any
mention of such mighty nations as the Bavarians and Alamanni
is remarkable. The word Swcefe occurs three times, but in each
case the tribe referred to is grouped with peoples dwelling on
the Baltic, and it seems that the North Suavi, not the great
Southern people, are meant.
(7) Out of thirty-one tribes and places which we can
identify all but two are connected with either the North Sea
or the Baltic : and these two exceptions are not far remote.
The deductions seem obvious. The geography is not that
of a traveller going from Angel to the Black Sea in the days
of Ermanaric. Neither is it that of one going from Angel to
Italy in the days of Alboin. Such a traveller must have
passed through the lands of the Alamanni and Bavarians :
Swcefe would have meant to him not the Baltic folk but the
mighty Southern tribe : he must have met with the Heruli,
Sciri, Turcilingi, Slavs, Avars, not one of which tribes is
mentioned. Of the races specially recorded by Paul as having
accompanied Alboin's Lombards into Italy — the Gepidae, Bul
garians, Sarmatians, Pannonians, Suavi and Noricans — two only
are known to Widsith.
Further, the names occurring in the unsuspected portions
of Widsith are systematic : they are not the casual accretion
of successive interpolators. The only non-Germanic names are
The Geography of Widsith 165
those of the border-peoples, which must have been known to
the most stay-at-home German. And even of Germanic tribes
those only are mentioned which might have come within the
view of one whose outlook was confined to the North Sea and
the Baltic — a tribesman situated be seem tweonum.
We have such a survey of Germanic geography and legend
as might have been made by a gleeman who drew his lore from
the traditions of the ancient Angel. From this point of view
everything drops into its right place : the knowledge shown
of the Anglo- Frisian tribes and of their legends, of their
immediate neighbours, and the more or less complete ignorance
of everything else, with the one exception of Gothic saga,
become at once intelligible.
I do not mean that the geography of the poem shows that
it was made before the Angles came to England : but that it
was made not long after, whilst the traditions of the continental
home were still fresh. How different the English outlook
ultimately became is shown in the geography of Germany added
by Alfred to the Orosius1.
When we turn to the suspected passages we find a casual
collection of names, most of them non-Germanic. Many of
these names, ^Elfwine, ^Egelmund, Hwala, may represent
genuine tradition: but taking them in the mass, they are
a jumble such as might have been made by some half-learned
and quite stupid person2. The phrase eastern of Ongle seems
definitely to point to an English origin3.
An examination of the geography of Widsith confirms us,
then, in the belief that Widsith cannot have any autobio
graphical basis, but that it represents an exceedingly early
form of traditional lore, and that certain portions, which can be
defined with some accuracy, are likely to be later interpolations.
The ease with which the passages containing the suspected
names can be disentangled, leaving nothing which does not
bear the stamp of great age, is remarkable.
1 See Appendix L.
2 Brandl supposes them the work of a cleric who had read Alfred's Orosius
(Pauls Grdr. 11, 1, 966-7. See Appendix L).
3 See note to 1. 8.
CHAPTER VI.
LANGUAGE AND METRE OF WIDSITH.
Dialect.
Widsith, like almost all Old English poetry, is written in
the normal Late West Saxon literary speech. But certain
words, chiefly proper names, show that the poem has been
transliterated from a non-W.S. dialect. Such are Breoca, Eatul,
Earmanric, HeaSo-, Seafola, Wald, freffSu-, geofum, meodu-,
sccecefi, sprecan1. Some of these forms, like Wald, though
commoner in Anglian, are not unknown in W.S. : but most
are quite impossible in a pure W.S. dialect. Equally significant
is the odd form Creacum, where we should expect Crecum.
For, if the scribe had been transliterating from an Anglian
original, he would have often had to change Angl. e to W.S. ea :
and he might easily have done this once too often, and have
wrongly written Creacum for Crecum, just as he would rightly
have written eac for ec*.
The dialect points, then, to a non-W.S., and probably
Anglian, origin for our poem : and this confirms what we
should expect from the place allotted to Offa, the greatest
of all Anglian heroes.
The language has been too much altered for it to give
much indication of date, except where a change involves
a difference in the number of syllables : in which cases the
1 Headen and wiolena rest upon conjectural emendation, and should
therefore hardly be counted : it is also conjecture that Eowan stands for
Eawan. Brandl quotes Deanum as non-W.S. : but we cannot be certain what
the original form of this word was.
2 Pointed out by Holthausen, n, 165.
Lcmguage and Metre of Widsith 167
metre sometimes preserves the earlier form, or more often
betrays what the earlier form must have been.
Syntactical Usages.
Attempts to determine the date of O.E. poems by observing
the usage of the demonstrative, se, seo, ]>cet, and of the weak
adjective, were made by Lichtenheld in 18721. Lichtenheld
examined the usage of Beowulf, Genesis, Andreas, Maldon and
the poems in the Anglo-Saxon Chronicle. He showed that,
taking Beowulf as a specimen of the earliest O.E. poetry,
Genesis and Andreas as intermediate, Maldon and the Chronicle-
poems as late, there is a progressive alteration in usage :
(1) se, seo, }>cet becomes more and more common2. This is
largely due to the fact that :
(2) in the older poems the weak adjective + noun occurs
frequently where we should now use the definite article : wisa
fengel, " the wise prince " ; whilst se wisa fengel is used only
when a demonstrative is needed, " that wise prince."
(3) Later, however, se, seo, \cet comes to be used in the
common and vague sense in which the definite article is used
in Modern English, so that we get with increasing frequency
the usage, definite article + weak adjective + noun3; whilst the
usage, weak adjective + noun, decreases4.
(4) The use of the instrumental, and particularly of the
instrumental of the weak adjective + noun, becomes more and
more rare5.
Lichtenheld also investigated (5) the use of the weak
adjective without noun (se wisa "the wise one"). His figures
here6 showed no very marked tendency either to an increased
or decreased use. These cases need careful examination and
discrimination, which, so far as space allowed, was given to
them by Lichtenheld7.
Lichtenheld's tests have since been applied to all the
longer O.E. poems by A. J. Barnouw8.
1 Z.f.d.A. xvi, 325, etc.
2 P. 332. • Pp. 335, 357-69. 4 Pp. 334, 375.
0 Pp. 326-7. 6 P. 331. i P. 352.
8 Textkritische Untersuchungen, 1902.
168 Widsith
That the use of the article becomes commoner as time goes
on in O.E. poetry is indisputable : how far we are justified in
using the statistics of Lichtenheld and Barnouw as criteria of
exact date is more open to dispute1.
Now Brandl2 notes in Widsith three examples of the use
of the weak adjective + noun (without the definite article).
Against this he alleges three instances of the use of the definite
article + weak adjective + noun, and accordingly pronounces, so
far as judgement is possible on such limited figures, for the
eighth century, and the period before Cynewulf.
But, as a matter of fact, only one of the three alleged
instances of the late usage, def. art. + weak adj. 4- noun, is
correct. The other two are examples not of the def. art. + weak
adj. + noun but of the def. art. + weak adj., a usage which needs
scrutiny and which, on the whole, may be an indication rather of
early than of late date3. These two examples are gestya \>a selestan
(1. 110), gesi\a }>a scemestan (1. 125). Now here the article is
fully justified, according to the oldest usage, and according to
what appears to have been the original force of se, and of the weak
adjective. Both Lichtenheld (p. 343) and Barnouw (pp. 42-43)
speak expressly with regard to this use of the article with the
superlative. Widsith here conforms to the usage of Beowulf,
where, as Barnouw points out, selest is employed for inanimate
things, se selesta for persons : husa selest, but magcfyegna Ipone
selestan. We certainly cannot regard it as in any sense a late
usage. We have here, then, in Widsith, not three examples of
early against three of late usage, but three of early against one
of late usage.
Widsith is too brief a poem to afford a satisfactory basis for
argument from statistics; on the other hand, its brevity will
permit of our applying all Lichtenheld's tests, and considering
all the circumstances under which the article is or is not used.
Such full examination is necessary, if we are to use the tests
with any degree of safety8.
1 See Sarrazin in Engl. Stud, xxxvm, 145 etc.
2 Pauls Grdr.(2) n, 1, 968.
3 Foster, Judith, 1892, p. 54 (Q.u.F. LXXI).
Language and Metre of Widsith 169
I. Use of se, seo, \>cet.
Se, seo, }>cet is used six times in Widsith (11. 36, 110, 118,
125, 127, 131): four times in undoubted, twice in suspected
passages. This is a very small percentage. We must of course
allow for the fact that 47 lines are filled only with lists of
names, and that in them we should hardly expect to find either
articles or adjectives. These lines should, no doubt, be dis
missed from the reckoning; but we are still left with six
instances in 96 lines, or one in every 16. The article is
found in Beowulf once in every 11 lines, more frequently in
Genesis, once in seven lines in the Andreas, about as often in
the Chronicle poems, once in four lines in Maldon1. So far,
then, as restraint in its use is a test of antiquity, Widsith is
more archaic than Beowulf, as Beowulf is more archaic than
Andreas and Andreas than Maldon.
When we examine the individual instances we find that in
five cases out of six the use of se, seo, ]>cet is justifiable : with
superlatives, ]>a selestan, ]>a scemestan ; with distinct demonstra
tive effect, of }>am heape " from that company of which I was
speaking," ]>ara monna modgast "bravest of those men just
mentioned," on ]>cere feringe " in my travels of which I have
been telling." Only in one instance )?a wloncan gedriht " the
proud company " have we the late, rather meaningless use of se
as the definite article. And this instance occurs in a passage
which we have already concluded, on grounds quite other than
those of grammar, to be probably a late interpolation.
II. Weak Adjective + Noun.
Of this early usage we get one instance in the suspected
lines (forman stye), two in the undoubted passages (wundnan
golde, sciran reorde). Excluding, as before, lines containing
proper names only, we have three examples in 96 lines, or, if we
prefer to take the undoubted passages only, two in 64: as it
happens, precisely the same proportion : a proportion greater
than the 80 instances in Beowulf, and much greater than the 25 in
Genesis, seven in Andreas, none in the Chronicle, two in Maldon*.
1 These are Lichtenheld's figures: Z.f.d.A. xvi, 332, 334.
. 2 Again I give Lichtenheld's figures : Z.f.d.A. xvi, 334.
170 Widsith
Here again, then, Widsith shows a usage more archaic than
that of Beowulf, and much more archaic than that of the other
poems.
III. Definite Article + Weak Adjective + Noun.
This usage, we have seen, occurs only once in Widsith, and
then in a passage which, on the ground of legend, looks like an
interpolation. The usage occurs 13 times in Beowulf1, 60 times
in Genesis, 12 times in the Chronicle-poems (a proportion equal
to 200 in Beowulf). If we take Widsith as a whole, this would
place it between Beowulf and Genesis. But if we are allowed
to take the undoubted passages separately, we find them entirely
free from this late usage.
IV. Weak Instrumental.
This occurs three times in the 96 lines, twice in the 64
undoubted lines : an equal proportion, and one larger than we
get elsewhere in O.E. poetry. Here again, then, we have an
argument making for very early date.
V. Definite Article + Adjective : ]>a selestan, ]>a scemestan.
This has been dealt with already. It is a usage from which
chronological argument cannot be drawn ; it is admittedly con
sistent with very early date.
I doubt whether we are justified in attaching very great
importance to figures so limited2. But, if we are to draw
inferences, then there is no doubt that the usage of the definite
article and of the weak adjective in Widsith points, not to the
eighth century and the time before Cynewulf, but rather to the
time before Beowulf, i.e. to the seventh century. Brandl's
statement to the contrary rests upon one of those slips from
which the work of the best scholars can never be absolutely
free.
1 Lichtenheld says 21 times (Z.f.d.A. xvi, 335). But eight of these are
really examples of def. art. + weak adj. (with noun in apposition), se goda,
m<eg Higelaces, not 'the good kinsman of Hygelac' but 'the good one, the
kinsman of Hygelac.' There remain 13 undoubted examples.
2 Cf. Sairazin in Engl. Stud, xxxvin, 146 etc.
Language and Metre of Widsith 171
Metrical Tests.
One of the most important contributions to the study of
O.E. made in recent years was the essay in which Morsbach
strove to prove that certain sound-changes, and especially the
loss of u after a long stem syllable, took place in England at
the end of the seventh century: that the metre of Beowulf
shows that the shortened form only was used, and that we must
therefore date Beowulf after 7001.
But the ground upon which this whole theory is built is
rendered infirm by the scantiness of the evidence. Two facts
stand out clearly : that in the Franks-casket inscription, dating
apparently from the end of the seventh century, the u is found
surviving (flodu for flod) ; that in charters dating from 692 or
693 the u is already lost2. The inference certainly is that
the u was lost at the end of the seventh century. But we
cannot be certain : flodu on the casket may be a deliberate
archaism3, it may be due to analogy, it may be a local
peculiarity.
Morsbach's test has been applied by Carl Richter4 to the
whole body of Old English poetry, with a thoroughness
accompanied by an absence of dogmatism deserving of the
highest praise. Richter dates Widsith6 after Beowulf, but
before Cynewulf : but in this he seems to have been largely
influenced by Brandl's argument as to the article6. A con
sideration in detail of the instances noted by Richter will,
I think, show that, even granting the accuracy of Morsbach's
data, and granting that their application to Beowulf proves
that poem to be later than 700, no similar result can be drawn
from their application to Widsith.
45a. Hro]>wulf and Hroftgar. "The substitution of *Hro!8-
garu would," says Richter, "make an impossible half line."
Even if this were so, it would not be quite conclusive against
a seventh century date. For, as Morsbach admits, the u in
1 Zur Datierung des Beowulf epos in the Nachrichten der k. Oesell. d. Wissen-
schaften zu Gottingen, 1906, pp. 251-72.
2 in silua quae dicitur uuidmundes felt. Kemble, C.D.S. i, 40.
3 Chadwick in the Trans, of the Cambridge Philological Society, rv, u,
116, 156.
4 Chronologische Studien, 1910. ° P. 92. 6 See above, p. 170.
Widsith
compound words was probably lost before the u in mono
syllables. The form Hrcfogar, then, is conceivable in the
seventh century. But I do not think that Hrcfogaru is
unmetrical here. On the contrary we should have a line of
a rare, but recognized type : Sievers " Expanded D, 9th sub
class1." The scheme is -*• x x | -^-6 x, and Sievers quotes seven
examples of it from Beowulf. For, Hrcfywulf being a proper
name, the second element may count either as secondary
accent or as a dip2. All that can be demanded is that such
a type, to compensate for its extra weight, should have double
alliteration. That condition being complied with there is no
further objection to be made. The earliest recorded half-line of
Germanic poetry is of this type8, complicated by an unaccented
syllable as " prelude " :
Ek Hlewagastiz Holtingaz.
The substitution of the early form *Hro$garu leaves us here,
therefore, with a satisfactory line.
This is hardly the case with Richter's second instance :
H7b, j/Egelmund and Hungar. If we substitute *Hungaru
here we certainly have an overweighted line, and one which
it would be difficult to parallel4. For, coming as it does in
the second half of the line, the extra weight cannot be carried
off by any double alliteration.
72a. leohteste hond. Substituting *hondu we should have
an uncommon line, but one which, as Richter admits, is not
unparalleled. It would be5 Expanded A*, subtype a. This
type usually has double alliteration: leohteste *hondu would
have but one parallel in Beowulf, Wiglaf wees haten (2602)B.
1 See P.B.B. x, 304. Cf. Beowulf (11. 712, 1790, 2462 ; 1426, 1440, 612,
232). Some of these verses, and the one under consideration, might perhaps
rather be regarded as expanded A types.
2 Dagegen haben die zweiten Glieder von Eigennamen (wie Beowulf, Hggelac),
nur einen schwacheren Nebenton, der nach Belieben des Dichters zur Bildung
eines besonderen Gliedes verwandt oder ignoriert werden kann. Sievers,
Hetrik in Pauls Grdr.($ n, 2, p, 30.
3 The dropping of the final unaccented syllables, which was completed about
the year 700, must have diminished the tendency towards the " expanded " type
of verse. Types which after 700 are rare may have been common when, with a
larger number of unaccented syllables in the language, there was more oppor
tunity for their use.
4 For wundor is to secganne (Beowulf, 1724) etc. cf. Sievers P.B.B. x, 255.
8 Also called expanded E, subtype 8 : Sievers, P.B.B. x, 310.
Language and Metre of Widsith 173
There is also the test of post-consonantal h before vowels.
This also is supposed to have been lost about 700. If Widsith
dates from the seventh century we should then have to read
and Wal[h]a rices x | -*• x | •*• x instead of and Wala rices x6x\-*-x:
an A line with prelude instead of a C line with resolution.
This would indeed give a less usual type of line, though one
not unparalleled : there are only four instances of the one in
the second half lines of Beowulf against 76 of the other1.
To sum up : there can theu, I think, be no dispute that
verse 45a gives us a satisfactory line, when we restore the
seventh century form. As to the other three, there might be
two opinions. But they all occur in those passages of the poem
which, on other grounds, we have marked as doubtful. If there
fore we grant these lines to be clumsy or impossible in a seventh
century form, we do not prove a post-seventh century date for
Widsith. We merely have a confirmation of the doubts which,
on quite other grounds, Kemble, Miillenhoff and Mb'ller have
entertained as to the originality of these very passages.
Finally, there are certain usages in which the earliest poems
fluctuate. For example, though few poems use early forms
exclusively, we can get some estimate of age from the proportion
of instances where the metre demands the Anglian mutated
form frega or the later contracted form frea. frea Myrginga
(Widsith, 1. 96), according to Richter, is an example of the later
form : to me it appears to be ambiguous ; we could quite well
scan fre[g~\a Myrginga, like dohtor Hrotyares*, a type of which
there are ten examples in Beowulf3: so that we can draw no
argument from this line. In the only other case where a word
of this type appears, }>eah }>e ic hy anihst, we must read anihst
in its earlier trisyllabic form anehist (x / x) if we are to make
the line scan at all. The proportions for early as against late
usage in this respect are then in Widsith 1 : 0, compared with
36 : 48 in Beowulf, 44 : 32 in Genesis A, 13: 5 in Daniel A4.
As to the treatment of syllabic liquids and nasals, we have
in Widsith, on Richter's reckoning, two examples of the earlier
1 P.B.B. x, 234, 244. 2 Cf. however Eichter, 11.
3 Sievers in P.B.B. x, 233, 255. * Eichter, 79.
Widsiih
against two of the later usage. The proportions of early to
late are in Beowulf 35 : 42, in Genesis A 17 : 73, in Daniel A
3 : 211.
These instances occur in the undoubted passages of Widsith.
It may be unwise to argue upon such limited evidence :
but if we are going to argue, the evidence again points to
Widsith being older than Beowulf, Genesis A, Daniel A — the
three poems regarded by Richter as the earliest.
Sentence Structure and Verse Form.
But there is another striking characteristic of Widsith,
which, fortunately, it has not been in the power of the scribe
to obliterate : the poet's habit of terminating his sentences at
the end of the line, with the result that the poem seems to
fall into irregular strophes of anything up to six lines in length.
Only three times does the full stop come at the half line, and
in each case in a passage which, on quite other grounds, has
been marked as interpolated. The exact number of stops in
the undoubted lines might be anything from twenty-four to
thirty-five : but they all come at the end of the line.
Now this is a most unusual feature in Old English verse,
where the stop occurs almost as often at the end of the half
line as at the end of the line. In two groups of poems only
do we find consistent and unbroken " end stopping"2. It occurs
in the long Riddle (No. XLI), in the Hymn (Grein-Wiilcker,
II, 2, 244), in the poems on the Day of Judgement and on the
death of Edward the Confessor, in the metrical version of the
Psalms almost without exception, and in the Kentish version
of Psalm Li. But these are all monkish effusions, written
under the influence of the ecclesiastical Latin : and we have
only to turn to the de die Judicii3, or the riddle of Aldhelm,
de Creatura, to see that the Old English poet is imitating the
structure of the Latin text he is translating, precisely as in the
1 Richter, 76.
2 Poems less than twenty lines in length are too short for purposes of
comparison, yet some confirmation of the argument below might be derived
from poems lite Ctzdmon's Hymn or the Leiden Riddle.
3 Bede, v, 634, in Migne, 1846.
Language and Metre of Widsith 175
Anglo-Saxon version of the Psalms the strophes correspond to
the verses of the Latin original.
But there is another class of poems to which this explana
tion does not apply : for if, on the one hand, we find this
" end stopping " in the monkish poetry, we find it on the other
hand in the most purely Germanic of all the Old English
poems : in the Charms, in the Rune Song1, in Deor, and here
in Widsith. This is not due to accident : for the tendency to
strophic form is emphasized in Deor by the refrain, and occasion
ally in Widsith and the Charms by the repetition of certain
words. Besides, the contrast is too striking to be accidental.
For, whilst the proportion of end-stopped to mid-stopped lines
is in Beowulf 385 : 282, in Genesis A 259 : 255, in the Elene
105 : 138, in the Andreas 221 : 112, it is in the Rune Song
29 : 0, in Deor 10 : 0, in the Charms 32 : 1 (or rather O2),
in Widsith (undoubted portions) 29^xtJ. To what, then, are
we to attribute this remarkable Difference?
That the earliest poetry of the Germanic tribes was lyric
and strophic is admitted. I believe the strophic character
of Widsith and of the Charms to be due to a survival of the
influence of this earlier Germanic tradition, precisely as the
strophic character of the monkish verse is due to the influence
of example, though in that case of an alien one. Of this
earliest lyric poetry we have several records in the historians3:
the fullest is the description of the Gothic lays sung in the
hall of Attila4. We are told how two minstrels came before
Attila, singing songs they had made of his victories and glory
1 Contrast with the Rune Song the enjambment of the Cynewulfian rune-
8 As some impartial criterion of what constitutes a full-stop is necessary,
I follow the punctuation of Grein-Wiilcker, where 1. 28 of the second Charm is
printed hal westu. Helpe }nn drihten. Yet a comma or colon would perhaps
he better.
I only count the three longer charms. If the shorter ones had been
included my point would have been unduly exaggerated, since, of course, a
number of short poems would tend to show a high percentage of "end
stopped" sentences.
3 Jordanes, c. 41, c. 49. Cf. also Sievers, Metrik in Pauls Grdr.fr) u, 2, 5.
4 Priscus (in Moller, Frag. Hist. Grcec. 1851, iv, 92). That the minstrelsy
and indeed all the civilization of Attila's court was Gothic seems certain.
Cf. Socin, Schriftsprache u. Dialekte, 1888, p. 8; Schutte, Oldsagn, p. 82.
Widsith
in war, and how the company were moved, even to tears. The
introduction of two singers together seems to postulate either
a choral lay or the singing of alternate stanzas ; the poem too
is encomiastic, not narrative. In both respects the minstrelsy
differs from that of Hrothgar's poet, telling in Heorot the story
of Finnesburh. Now, when Widsith sings before his lord, it
is with a second minstrel : the two together sing one song
(for the company say they had never heard, not better songs,
but a better song) ; the song too is not a narrative, but a song
of praise, the composition of the singers themselves. The
coincidence may, of course, be accidental, and encomiastic and
narrative poetry flourished, we know, side by side ; the names
in Widsith are themselves evidence for the existence of many
narrative lays. Yet it is worth noting that the singing before
Widsith's lord reminds us rather of the hall of Attila than
of Hrothgar1.
And it can hardly be accident that four poems, the subject-
matter of which carries us back to the most primitive period,
should agree in the retention of a primitive form. They may
have been modernized and interpolated : we know that the
Rune Song* and the Charms have been largely rewritten : but
all four seem to have their roots in an age more primitive
than that which rejoiced in the rhetorical interlaced periods
of the Beowulf or Genesis : an age when the tendency, inherited
from the strophic poetry, to make the sentence end with the
line, was still strong.
And here, as elsewhere, we find a distinction between the
parts of Widsith : the Introduction, which on other grounds we
have thought to be later, shows the enjambmenfc familiar in
Beowulf: whilst in 11. 17 etc. which, on other grounds, we have
found to be exceedingly early, we have, not merely "end stopping,"
but definite and regular strophic form3.
1 For the importance of Widsith as illustrating the duties and position of
the scdp, see Anderson, passim, and Kohler in Germania, xv, 37 etc.
2 Cf. B. M. Meyer in P.B.B. xxxn, 76. 3 Cf. Neckel, pp. 1-21.
CHAPTER VII.
SUMMARY AND CONCLUSION.
THE passages in Widsith which show traces of being later
additions, and which were on that account rejected by the
earlier critics, have been enumerated at the end of Chapter iv.
A survey of the geography of the poem1 has emphasized the
contrast between the stamp of definite locality borne by the
undoubted portions, and the chaos of names found in the
doubtful passages. A study of the grammar of the poem2
has shown that such examples of late usage as we do find
occur in just those passages which the earlier critics, on
grounds quite other than those of metre or language, had
suspected of being later additions8. The poem as a whole,
and still more the undoubted portions, taken by themselves,
we find to be more primitive in grammar and metre than
the poems usually regarded as the oldest in English literature,
such as Beowulf and Genesis A*. Above all, the habit of always
concluding the sentence with the line, never with the half line,
dissociates these undoubted portions of the poem from Old
English epic poetry, and connects them with a small group
of poems which, in subject matter, all seem to go back to a
more primitive period than that of the courtly or learned epic,
as we know it8.
Reason has been shown for believing that these undoubted
portions of the poem fall into two sections6, originally distinct,
the Catalogue of Kings, and the lay of Ealhhild and Eormanric
1 Chap. v. 2 Chap. vi. » pp. 169, 173.
Vpp. 170, 174. B pp. 175-6. « pp. 133-5.
c, 12
178 Widsith
which we may regard as the essential Widsith, Widsith, alike
on grounds of legend1 and of geography2, cannot be the work
of a contemporary of Eadgils and Ealhhild who really visited
the court of Eormanric. The Catalogue of Kings is older
than Widsith proper8, yet on account of the names it contains
it can hardly be earlier than the* middle of the sixth century,
and may be considerably later. Widsith seems to belong to
a period later than this, but earlier than Beowulf or Genesis :
that is to the seventh century.
This too is the date which has already been approved by
those who have studied the poem chiefly from the point of
view of heroic legend4 and of geography6: it has been widely
accepted, but not universally, because the view has hitherto
been entertained that the language and metre of Widsith
pointed rather to the eighth than to the seventh century.
This view has been shown to rest partly upon a miscount,
partly upon a failure to differentiate from the rest of the poem
those parts which (on grounds not grammatical or metrical)
had been already suspected of being later interpolations.
It has still to be shown that interpolation and amalgamation
of the kind postulated are inherently probable. For, as has been
urged above, it is not enough to show that certain passages are
inconsistent with others, in order to dismiss one or other as an
interpolation. A reason which can have prompted such inter
polation should be forthcoming6. We only escape from one
difficulty into another, if we explain an inconsistent text by
assuming an irrational interpolator.
Now, if a poem existed telling of the journey of a Myrging
minstrel to Eormanric, and enumerating the tribes and heroes
he had met, and if, as is most probable, a mnemonic catalogue
of tribes and heroes existed of the kind we find so often in
kindred literatures7, the tendency to amalgamation would be
strong. If these poems did not contain the name of ^Elfwine,
then, when ^Elfwine passed from history to the same plane
1 p. 142. 2 pp. 163-5. 8 pp. 134-5, 150, 176.
* e.g. Symons. See above, p. 150. 5 e.g. Much, 65.
B See above, pp. 6, 144. 7 See above, p. 6.
Summary and Conclusion 179
of heroic poetry as Eormanric, the temptation to add some
lines mentioning the generosity of ^Elfwine would also be
strong. An introduction binding the somewhat inconsistent
Catalogue of Kings1 with Widsith proper2 and explaining who
Widsith, Ealhhild and Eormanric were, would become necessary,
more particularly when the Eormanric cycle was beginning to
be less generally familiar.
Again, our poem owes its preservation, where so much has
been lost, to the fact that it interested a monkish scribe,
probably because of its encyclopaedic geographical information.
But the world of this scribe was very different from that of the
tribal bard who first devised the lay. Could a man of ancient
days have been really travelled, he would ask, if he did not
know of Alexander, had not seen Medes and Persians and
Hebrews ? Hence interpolations, which are but a necessary
corollary of the new civilization which had been superimposed
upon the old Germanic life. There may have been a more
definite motive : the scribe may have wished to bring up the
number of tribes to 72, the supposed number of the nations
of the earth3 : or the poem may have been transmitted to him
as an imperfect one, and he may have been filling the gaps as
best he could. The latter hypothesis seems the more probable
one : for there are incoherencies in the poem which may well
be the result of something having been lost.
Such a development seems then to be likely enough on
a priori grounds : whether the evidence produced in its favour
has been sufficient the reader must judge. The strength of
such evidence varies, of course, in the different cases: the
1 It would be difficult to say how far the parallels between the Catalogue
of Kings (11. 18, 20, 21, 44) and Widsith proper (57, 76, 69, 61) are the cause,
and how far the result, of amalgamation. The collocation of Huns and Goths
is part of the common Germanic tradition, found also in the Elene, and in
Icelandic ; its repetition need not therefore be due to imitation ; 1. 76, on the
other hand, is probably the conscious imitation of an interpolator.
8 The statement that the Myrging Widsith started from Angel is strange
(see note to 1. 8). It may point to some details of the legend, or some lines
of the poem which have been subsequently lost. If not, it looks like a
conscious attempt to harmonize the Anglian patriotism of 11. 38-44 with the
Myrging patriotism of 11. 93-96. LI. 84-85 may be an interpolation having
the same object, (See note to 1. 85.)
8 B. Michel in P.B.B. xv, 377 etc. : cf . Bojunga in P.B.B. xvi, 545 etc. ;
Brandl in Pauls Grdr.(%) n, 1, 967.
12—2
Widsith
spuriousness of the "Biblical" passage rests upon grounds
of outlook, geography, and metre which seem to me over
whelming : the spuriousness of the ^Elfwine passage rests upon
one chronological and one grammatical detail, neither of which
appears to me quite conclusive.
h
But much more important than these controversies as to
the structure of Widsith are the conclusions we can draw from
the poem regarding the ancient poetry of our race.
The place of Widsith in the history of Germanic
poetry.
In attempting to estimate the value of the Old English
heroic poetry, allowance is too often not made for the frag
mentary way in which that poetry has been handed down to
us. We have not, as we have in Homer, that which the con
sidered judgement of the race thought most worthy to survive
in its national poetry. The story of Beowulf is preserved, not
because it was better or more popular than other stories, but
because it happened to get written down (by no means a
criterion of merit, in an age when there was little sympathy
between the monk and the minstrel) ; because this copy hap
pened to get transcribed before it was destroyed ; because, in
the Middle Ages, it happened that nobody was sufficiently
energetic to cut up the Beowulf MS in order to bind more
intelligible books ; because at the dissolution of the monasteries
it happened not to be used for packing paper; because it
happened not to blaze up in the great Cottonian fire. It must
not then be taken for granted, without examination, that Beo
wulf is a fair specimen of the old heroic poetry, still less that it
is the best specimen.
A comparison with the extant fragments of the older heroic
verse, with later poetry like Maldon, and with those poems in
which biblical stories are told in the old epic style, shows us
that in style at least Beowulf is a noble representative of its
class. But is it so in plot ?
Monstrous and childish elements, belonging to the youth of
a nation, are often found surviving sporadically in the poetry
Summary and Conclusion 181
of a more mature age : the Cyclops and the No-Man story, for
example, have kept their place in the most polished of all
epics. Now, if the story told by Odysseus in the hall of
Alcinous were all that remained to us of Homer, we should
be in much the same position in judging the Greek epic as we
are when we try to judge of the Old English heroic poetry from
Beowulf. We should be puzzled by the obvious incongruity
between the childish story, and the epic dignity with which
that childish story is treated. And, just as we might have
been helped to a truer perception of a lost Homer by the
briefest summary of the fight in the hall of Odysseus, or
by even such a travesty of the xxivth book of the Iliad
as Tzetzes has left us, so we can to some extent reconstruct the
matter of the Old German lays from Paul the Deacon or from
Widukind, from the Elder Edda and the sagas.
Herein consists the real value of Widsith : it shows us what
was the stock-in-trade of the old Anglian bard : and if Widsith
be of composite origin, this only makes the evidence more
representative. Read in a sympathetic mood, by anyone who
has taken the trouble to acquaint himself with such of the old
stories as time has left to us, Widsith demonstrates the dignity
of the Old English narrative poetry, and of the common Ger
manic narrative poetry of which the Old English was but a
section. The conquerors of Europe were not primarily inter
ested in inarticulate water-demons or fire-drakes. The stories
told in their lays were amongst " the best tales sorrow ever
wrought." And we know from Beowulf that the style was not
unworthy of such matter.
We find in Widsith no allusion to Beowulf: no reference to
fights with giants or dragons : Wada, the only clearly super
natural character mentioned, has been rationalized as the chief
of a historic tribe1. The supernatural had a place in Old
English heroic verse : but, as in the Greek epic, this place was
a subordinate one. If we pass once more hastily in review
these old Germanic tales, we shall see that what chiefly
1 For evidence that the Hcelsingas are historic see note to 1. 22. Breca may
have been originally mythical: bat he is grouped with historic kings. See
above, p. 111.
Widsith
interested the poet was character : the behaviour of his hero
or heroine under the stress of conflicting passions. The story
of Eormanric tells how, stirred by an evil counsellor, a great and
aged king was moved to take a wild revenge upon his young
bride, and how vengeance befell him ; the tale of Hagena and
Heoden tells how mischief was wrought between a young chief
and the father of his betrothed, so that no reconciliation could
be made ; similar, in its main outline, is the story of Sigehere
(Sigar, Hagbard and Signy), save that here, if Saxo has told
the tale aright, the poet was interested rather in the lovers
than in the father; in the story of Ingeld and Hrothgar we
have the same tale once more, but again with a difference, for
here it is the mischief-maker himself who attracts the poet's
interest, the eald cesc-wiga whose loyalty to the old time is
such that he cannot allow the ancient grudge to be forgotten.
The story of Guthhere was one of retainers falling around their
lord ; in the lay of Hnaef and Finn the retainers at last take
quarter, and heartburnings and new sorrows ensue ; in the
tale of Hrothgar and Hrothulf a noble chief usurps the throne,
not without injustice and perhaps treachery to the sons of his
benefactor, and in the end falls himself by treachery. Occa
sionally, though rarely, the tragedy is relieved. In the story
of Wermund and Offa, the grief of the old king, whose heir is
unworthy to follow him, is turned into joy when, in the
moment of peril, that heir proves himself worthy. In the
greatest of all the stories, that of ^Elfwine at the court of
the Gepid king, the ties of hospitality overcome the passion for
revenge.
Of course we cannot be certain that in every one of these
cases the name conveyed to the Anglian minstrel in the
seventh century just that meaning which has been here indi
cated : we cannot say, for example, how much story had as yet
collected round the name of ^Elfwine. But the general
deduction to be drawn from these names is not open to doubt.
It is that the old poets had the right tragic feeling, the sane
and just outlook. The situations are those which we find later
in the Icelandic sagas; in Othello and Romeo and Juliet, in
Henry IV and Macbeth.
Summary and Conclusion 183
The place of Heroic Poetry in the Dark Ages.
And, apart from questions of poetic form and literary
history, the fragmentary records of the old heroic poetry
have their value in helping us to estimate the meaning of
the "Dark Ages."
With certain notable exceptions the events of Western
Europe during these ages are told by contemporaries only
in dry annals or very meagre histories, which chronicle the
facts, mostly bloody and often disgusting, but leave us in
the dark as to the motives of the men who did the deeds.
Sometimes the history of Western Europe seems to be only
one monotonous sordid tale of lust and bloodshed. Yet this is
an illusion. The acts of fratricide and treachery which caused
an upheaval are just the things which find their way into the
brief annals ; the daily life, the long years of faithful devotion
of the retainer to his lord, are taken for granted. Yet without
such faithful devotion the retention of the throne during many
centuries in the hands of one family, Amal, Bait, Merowing,
Iceling, would not have been possible. We obviously need
some corrective to the annalists.
The motives of the leading figures in the history of these
ages can seldom be accurately measured : it is still more diffi
cult to estimate the feelings of the rank and file. But the
fragments of the old heroic poetry show us at least what kind
of men these retainers, the " swaese gesithas," wished to be.
The story of England in the days of Ethelred the Unready, as
told by chroniclers and homilists, is a sorry record of defeat.
We are ready to exclaim with Falstaff "Is there no virtue
extant ? " till the poem of Maldon makes us feel ashamed of the
estimate we have formed of tenth century England. Or to
take a rather earlier and perhaps healthier period. The eighth
century, to judge from the Anglo-Saxon Chronicle, seems a
welter of internecine strife, treachery and cruelty ; yet the men
among whom the song of Beowulf was loved, however given to
deep drinking and hard fighting, were not, in the main,
treacherous and cruel. It is the fallacy of the brief, unex
plained annal. Stated in bare annalistic form, the deeds of
184 WicMth
Shakespeare's noblest men, of Brutus, Hamlet, or Othello, would
appear as the acts of a gang of gaol-birds. Had we more of
the old lays, whether English, Gothic, Old Saxon or Frankish,
this impression of cruelty would probably be softened; we
should find that deeds of murder and treachery did not meet
with approval when songs were sung and talk was free in the
mead hall, after the day's hunting.
Only too often we hear in the annals of the guardian uncle
or retainer who supplants his young king. What Gothic singers
had to say about such things we learn from Cassiodorus' allusion
to the song of Gensimund. Gensimund, a valiant chief, though
not by birth an Amal, was adopted into the royal house. After
the king's death the throne was offered to him, but he refused
it, remaining faithful to the young princes, so that he is famed
in song and will never be forgotten so long as the Gothic name
lasts1.
Of Beowulf almost exactly the same is told us. He was
received by his king, and treated no whit worse than the king's
own sons2. When the old king and his sons were dead, the
last widowed queen offered the kingdom to Beowulf : " she
trusted not to her son that he could hold the throne against
foreign foes, now his father was dead. But none the more
could the helpless folk in any wise prevail upon the prince
[Beowulf] that he should be lord over Heardred and accept the
kingdom : but he supported him amid the folk with friendly
counsel, love and honour, till he grew older and ruled over the
Weder-Geatas3."
1 Heinzel (Hervararsaga in W.S.B. cxiv, 497) suggests that the faithful
Gizur of that saga is a survival of Gensimund. Mullenhoff (Z.f.d.A. xn, 254)
sees in Gensimund the historic forerunner of the Hildehrand of later story.
Mullenhoff (as subsequently Scherer) assumes that Gensimund has a place in
Gothic history, as the guardian of the father and uncle of Theodoric the Great.
He would thus come very near to the Master Hildebrand of High German
tradition. But the insertion of Gensimund into Gothic history at this point is
a mere conjecture of R. Kopke (Die Anfange. des Konigthums bei den Gothen,
Berlin, 1859, p. 141) to which further currency was given by Dahn's apparently
independent hypothesis (Konige der Germanen, Bd n, Miinchen, 1861, pp.
60-61). The difficulties of this chronology are shown by Hodgkin (Italy, in, 9).
We have no sufficient grounds for making Gensimund an historic figure, though
he may well have been such. W. Scherer (Review of Heyne's Bedvulf in the
Zeitschrift fur die osterreichischen Gymnasien, xx, 1869, p. 95) seems to have
been the first to draw attention to the remarkable parallel between Gensimund
and Beowulf.
2 2435, etc. s 2370, etc.
Summary and Conclusion 185
Nowhere do we find the judgement of the mead-hall ex
pressed better than in the change undergone by the story
of the treachery of the Frankish Theodoric. The Theodoric of
history lured his old foe Irminfrid of Thuringia into his power,
with promises of safe conduct and with rich gifts. Then, as
they were talking one day on the walls of Ziilpich, Irminfrid
was pushed over, and killed. Who actually did the deed was
never known : Theodoric was suspected of having insti
gated it1.
But poetry did not allow Theodoric to enjoy the fruits of
his treachery. In the place of the unknown assassin was
substituted a demi-god : that Iring who had left his mark
across the heavens in what we call the Milky Way. We
wonder what that figure is doing in the character of traitor :
till, when Theodoric disclaims the crime, Iring rises to the
height of his daemonic power, slays the perfidious king,
humiliates him even in death ; and then making his way
through the midst of the royal retinue, vanishes.
" The greater part," said Sir Thomas Browne, " must be
content to be as though they had not been ; to be found in the
register of God, not in the record of man. Diuturnity is a
dream and folly of expectation." Yet in the old heroic poetry
we get a glimpse of the thoughts of those men whose unre
corded lives and deaths have done more towards the building
up of Europe than have the intrigues and quarrels of their
lords. This should render sacred not only every recorded line
of the old poems, but every paraphrase and every allusion.
1 Sed quis eum exinde deiecerit ignoramus ; multi tamen adserunt Theudorici
in hoc dolum manifestissime patuisse. Gregorii Hist. Franc, p. 116.
TEXT OF WIDSITH, WITH NOTES.
A very accurate facsimile of the Exeter Book was made for the British
Museum, and checked by Sir Frederic Madden in Feb. 1832 (Add.
MS. 9067). It was upon this facsimile that Kemble based his text, but
only after comparison with the transcript of the Exeter Book which had
been already made by Thorpe, who, however, did not publish till 1842.
Mr Anscombe's theory (Anglia, xxxiv, 526-7) that Thorpe derived his
text from Kemble is disproved by Kemble's express statement, and by
the fact that Thorpe's text is much the more correct of the two.
Later editors based their texts on Kemble and Thorpe, either directly
or at second hand, till Schipper again collated the Exeter Book in 1870-1,
and published his results in 1874. Wiilcker again collated the MS. for his
Kleinere Dichtungen (1879 ; 1882). The text in his edition of Grein's
Bibliothek (1883) is even more accurate, and may be the result of a
second collation. Moller, who published in the same year, and all
subsequent editors, based their texts upon this.
Some details as to the relations of the different editions of Widsith to
one another, and of the British Museum facsimile to the MS. at Exeter, are
given by the present editor in Anglia, xxxv, 393-400.
I have twice examined the MS. and noted the marks of length, and
one or two other details not recorded by Wiilcker : but there is nothing
of importance to be added to his collation of the MS.
A collation of the seven chief texts of Widsith was appended by
Wiilcker to his text of the poem, issued in 1883. In a volume specially
devoted to Widsith it seemed right to go somewhat more into detail. The
first and second editions of Kemble have accordingly been distingxiished
(Kj and K2). They differ considerably, Kemble having incorporated into
his second edition many important emendations. As Leo founded his
text upon Kemble's second edition, with a general acknowledgement only,
credit for Kemble's elucidations of the text has sometimes been given to
later editors. By collating the edition of Ettmviller published in 1839,
and Lappenberg's commentary of the same year, the important part
played by both these scholars in the interpretation of the text of Widsith
is made more clear.
The texts given by Ebeling, Schaldemose, Klipstein, Moller, and Kluge
in his Lesebuch have also been consulted. The text of Ebeling is exceed
ingly bad ; he follows Ettmullerd) but with many misprints and many
wanton alterations, whilst Klipstein derives direct from Thorpe (Cod.-
Ex.}. Schaldemose gives the MS. readings, recording the conjectures of
Kemble, Leo, Ettmiiller in his footnotes. He generally adopts these
conjectures in his translation. Corruptions have, however, crept into
his text (e.g. 11. 33, 93, 98, 118). Accordingly the readings of Ebeling,
Klipstein and Schaldemose have not usually been recorded. Nor have
188 Widsith
Holler's transpositions been noted, as they can be clearly understood
only by consulting his complete text. The texts of Holthausen and
Sedgefield, which appeared whilst this edition was in the press, have
also been compared.
A number of the readings of the earlier editors are due to the elemen
tary condition of Old English studies in their time : such for example as
that of Ettmiiller and Schaldemose in 1. 134 ]>enden he here [for her] leaf aft
" so lang er Heerfahrt liebet " : " mens Feide ban elsker." Whilst fully
recognising how excellently the earlier editors did their work, I have not
felt compelled to carry respect for them so far as always to record these,
and similar obvious blunders, which they could not have made had they
possessed the apparatus of dictionaries and grammars which we now have.
Neither have readings been recorded which are misprints or oversights.
The earlier editors, Kemble, Leo, Ettmiiller, attempted to normalize the
text, writing or suggesting on for an (3rd pers. plu.), i for ie, h for g, est
for ast, eo for ea, se for e. In most cases it seemed unnecessary to record
again wanton alterations of this kind, as they will be found in the appa
ratus criticus of Grein-Wulcker. In the few cases where my collation did
not agree with that of Wiilcker, I have verified the reading.
WIDSID MADOLADe, wordhord onleac,
se be [monna] maest maet/ba ofer eorban,
folca geondferde: oft he [on] flette gebah
mynelicne mabbum. Him from Myrgingum
Widsith spake, unlocked his store of words, he who of all men had
wandered through most tribes and peoples throughout the earth : oft in
hall had he received the lovely treasure. His race sprang from the
1. WidstfS. A proper name, that of the supposed traveller : not, as the earlier
editors, before Thorpe (1842), take it : " the long journey " spoken of by him
"who wandered through most tribes." Wiilcker (Kl. D. 164) also supported
this rendering : bnt in his Grundnss (328) he regards WidstiS as a proper name.
The name is recorded as having been borne in historic times (Uidsith, Liber
Vitae, 179). Cf. Koegel, Ltg. i, 1, 138.
2. se \e monna mcest mceg]>a ofer eor]>an. The MS. reads se \>e mast mcer\>a
ofer eorban ; so Kj : K2 conjectured se \e mast fandode mceg\>a ofer eor\>an : Leo,
Ettmiiller d) and Schaldemose adopted fandode but kept mcefya, which they
rendered "das was geriihmt wird iiber die Erde," " Mannruhm " etc.: mcest
gemunde mcer&a, Ettmiiller (2) andBieger : mcest mette mcerSa, Thorpe (Cod. Ex.):
mast mcegfta mette, Thorpe (Beowulf).
se \e monna mcest mceglpa ... is Grein's reading ; it was at once endorsed by
Miillenhoff (Z.f.d.A. xi, 275) and later by Moller, Gr.-Wiilcker, Holt, and
Sedg. Eluge, however, retains meer\>a, being probably influenced by the parallel
passage in Beowulf (2645)
/orfcan he manna mcest mceriSa gefremede.
For the construction, cf. also Judith (181)
\e us monna mcest mofftra gefremede
sarra sorga.
on flette. on supplied by Grein, followed by all later editors.
4. mynelicne. a?ra£ \ey6/j.(vov. Not " memorable " as Thorpe (Cod. Ex.)
following Ettmiiller's " Erinnerungs Kleinod": but "lovely," " pleasant," cf.
Icel. munlegr, " pleasant," O.S. munilika magcfc, " a lovely maiden."
5. Him from Myrgingum cet>elo onwocon. The MS. has hine from Myrgingum
ce\ele onwocon, concerning which Conybeare remarked ' ' it may imply that the
nobles of bis own country had encouraged him to travel, as appears to have
Text (tt. 1-8) 189
5 aej^elo onwocon. He mid Ealhhilde,
fselre freojmwebban, forman sij?e
HreScyninges ham gesohte
eastan of Ongle, Eormanrices,
Myrgings : it was with the gracious lady Ealhhild that he first, from
Angel in the East, sought the home of the Gothic king Eormanric, fell and
faithless.
been the case with Gunnlaug. See Gunnlaug Sag. p. 96." [The reference is to
the Copenhagen edition of 1775. But the passage probably does not bear the
interpretation put upon it. Yet the travels of the poets Gunnlaug and Widsith
offer an interesting parallel.] Kemble, Leo, Ettmiiller and Thorpe also keep
the MS. reading. This, however, is impossible, for onwacan is intransitive.
Thorpe (Cod. Ex., notes) suggested, though he did not read, him : "from him
among the Myrgings nobles sprung," comparing Beowulf, 112, and the mention
of cnosl below. Klipstein follows Thorpe's conjecture, which is possible, but
not likely, for cnosle in 1. 52 probably means kindred, not offspring. Him from
Myrgingum aftelu onwocon is Grein's reading, and has been followed by later
editors (aebelo, Bieger, Grein-Wulcker, Kluge, Holt., Sedg.).
5. Ealhhilde. See Introduction, pp. 21-28.
6. falre freobuwebban. Fcele in this connection is a hackneyed epithet : feele
friSowebba (Elene, 88) : fcele freoftoscealc (Gen., 2301, 2497), fcele freoftuweard
'(Guthlac, 144), of. Holler, V.E. 32-3; Grimm, Andreas u. Elene, 143-5;
Lawrence in Mod. Philol. TV, 350-1.
Freobuwebbe was rendered " Fridensweberin " by Leo ; so Ettmiiller, Thorpe,
etc. But it must not be supposed that Ealhhild goes to the court of Eormanric
to negociate peace. The idea conveyed by freobuwebbe is rather that of
domestic peace than of diplomatic service ; cf. Lawrence in Mod. Philol.
iv, 350.
forman sibe, Gummere(i) translates " once." But it probably means more
than this. In Beowulf (716) where forma sfS occurs in a similar context,
it is emphatic enough :
Ne wees beet forma sift
bat he Hrobgares ham gesohte.
So Gummere(2), "first."
7. Hreftcyninges. See Appendix, Note H : The term Hreedas applied to the
GotJis. Holt, regards as a common noun: glorious king, "Ruhmkdnig."
But cf. Frescyning (Beowulf, 2503).
8. Ongle. The old home of the Angles, the Angulus of Bede, the Angel of
King Alfred. Now, in the second century, the Goths were dwelling on the
banks of the lower Vistula, due east of Angel. By the third century they had
left their old home, and were soon on the shores of the Black Sea, whilst by the
sixth century they were mainly in the south and west of Europe. Critics,
reading eaxtan of Ongle as a statement that the Goths dwelt east of Angel
(Grimm, Heldensage, 18, 19 ; Leo, 76 ; Boer, Ermanarich und Dietrich, p. 12 ;
cf. Weiland, 15 ; Gummere in M.L.N. iv, 210), have seen in this a proof of the
early date of the poem (Miillenhoff, D.A. n, 99 ; Klipstein, 11, 422 ; Moller,
V.E. 32; Weiland, 15 ; Koegel, Ltg. i, 1, 179). The words, however, cannot
mean this : " The home to the east of Angel " would be ham be eastan Ongle.
In the course of a controversy with Sarrazin, Sievers examined elaborately
the syntax of the O.E. adverbs of place with verbs of motion (P.B.B. xn, 168,
etc. ; also, less important, xi, 361 ; cf. Erdmann, 48). In Pauls Grdr. d) i, 408,
he shows the bearing of this upon Widsith. Eastan of Ongle means " from the
east, from Angel," and was apparently written by a resident in England,
referring to the old Anglian home in the east. It therefore tends to prove the
later date of, at any rate, this introduction. But the whole thing is puzzling.
Why should a Myrging bard start from Angei ? See Introduction, p. 165, 179, note.
190 Widsith
wra)>es wserlogan. Ongon J?a worn sprecan:
10 " Fela ic monna gefrsegn msegjmm weald an ;
sceal beod[n]a gehwylc beawum lifgan,
eorl sefter obrum eSle rsedan,
se }>e his beodenstol gebeon wile !
85 a || fara wses [HJwala hwile selast
15 ond Alexandreas ealra ricost
He began then to speak many words :
" Of many men have I heard, ruling over the nations. Every chieftain
must live virtuously (one lord after another, ruling his land), he who
desires his throne to nourish.
Of these was Hwala for a time the best, and Alexandreas most mighty
9. wrabes warlogan. The same expression is in the Andreas (613) applied
to the Devil. Thorpe assumed a gap : "Here some lines are evidently wanting,
although there is no hiatus in the MS., as the words wrfyes wserlogan cannot
apply to Eormanric, the object of the poet's praise." But this assumption is
groundless. See Introduction, pp. 34-36.
worn. Many words ; cf. Beowulf (530).
11. beodna. MS. beoda, so Kd). beodna, the conjecture of K(2), was adopted
by Leo, and has been accepted by all later edd., except Thorpe (Cod. Ex.) and
Ettmuller(2).
12. eorl after obrum. See Leo (87). Moller (V.E.) followed by Ten Brink
(539) transposed verses 11, 12, to the great improvement of the sense.
14. Hwala. MS. Wala, followed by K(i)(2), Leo, Schaldemose. Thorpe and
Eieger attribute to Kemble the emendation Hwala : but it seems to have been
made by Ettmuller (J. It is clearly right, and has been adopted by all later
edd. Hwala, son of Beowi (Bedwi) son of Sceaf, is found in the West Saxon
pedigree in three MSB. of the Chronicle (Cotton Tiberius A vi, B i, B iv : see
entry for year 855, death of ^Ejjelwulf).
That Kemble should have overlooked so obvious a correction here is strange,
for whilst editing Widsith he was also working at the O.E. genealogies, and had
pointed out that Wala in the later versions of these genealogies should be read
Hwala : " Wir werden hernach sehen dass die Lesart Hwala allein die richtige
ist, wenn wir die herrschende Alliteration in diesen Listen betrachten."
See Kemble, Ueber die Stammtafel der Westsachsen, Munchen, 1836, pp.
11-13, 26.
15. ond Alexandreas. Obviously Alexander the Great : though Grein would
read ond as a preposition, " among the Alexandrians," and Haigh, in his vain
attempt to make all the heroes of Widsith contemporary, interpreted the name
as a reference to one of the sixth century Alexanders mentioned by Procopius.
Holt, suggests Alexandras.
Mullenhoff, followed by Bieger, condemned 11. 14-17, on the ground that
they interrupted the sequence, for after the mention of Ermanaric in 1. 8 we
should expect the catalogue to begin among the Goths and their neighbours :
he therefore supposed 14-17 interpolated by the same hand as 11. 75-87 ; they
"show the same combination of national tradition [Hwala] with idle monastic
learning" [Alexandreas, Alexander the Great, with the name assimilated to
that of the saint Andreas]. Other edd. (e.g. Ebeling) condemn only 1. 15
(Alexandreas). The fact that the scribe of the Exeter Book, or perhaps an
earlier transcriber, did not know who Hwala was, and misspelt his name, shows
that Hwala had been forgotten, and is therefore against the theory that line 14
is a late, monastic, interpolation. Nor need the reference to Alexander be the
result of monastic learning. That the Alexander story was well known
throughout western Europe is proved by king Alfred's reference to Nectanebus
Text (II 9-19) 191
1 6 monna cynnes ond he rasest gebah,
]?ara J>e ic ofer foldan gefrsegen haebbe.
-(Etla weold Hunum, Eormanric Gotum,
Becca Baningum, Burgendum Gifica.
of all the race of men, and flourished most of those of whom I have heard
tell throughout the world....
(Orosius, ed. Sweet, 1883, E.E.T.S. p. 126), by the Macedonian descent attri
buted to the Franks (Otfrid, Evang. i, 1, 87-91) or Saxons (Widukind, i, 2),
and by the Old English translation of Alexander's letter to Aristotle. The
mediaeval Alexander saga has its root in popular traditions going back to the
third century of the Christian era (see Notice des Manuscrits contenant I'histoire
fabuleuse d'Alcxandre le Grand par J. Berger de Xivrey, in the Notices et
Extraits des Manuscrits de la Bib. du Eoi, Paris, t. xm, 1838, p. 178 : Spiegel
Die Alexander sage, Leipzig, 1851, p. 2). Alexander may therefore well have
been known by name among German tribes at a very early date. According to
Jordanes Ermanaric and Alexander had been coupled together : Hermanaricus
...quern merito nonnulli Alexandra Magno comparavere maiores (ed. Mommsen,
xxm). So, though 11. 14-17 are probably later, we cannot be certain.
18. Mtla. Kluge has pointed out that this form of the name corresponds to
the O.N. Atli, as against the M.H.G. Etzel, O.H.G. Ezzilo (by i-umlaut from
Attilo). The English and Norse forms come from a syncopated *Atlo, pre
sumably the Low German form about 500 A.D. See Engl. Stud, xxi (1895),
447. But the form Etla also is found in England.
Hunum.... Gotum. A rough approximation to geographical order can be
traced throughout this catalogue: 11. 18, 19 deal with the tribes and chiefs
of Eastern Germany. Ostrogoths and Huns were not only neighbours, but
closely associated in story both as allies and as foes. Hence they are again
coupled below (1. 57) and twice in the Elene : Huna leode ond Hreftgotan, 20;
Huna ond Hrefta here, 58. So in the Icelandic Lay of Hloth and Angantyr :
Ar kv6$o Humla Hunom rctfSa,
Gitzor Gautom, Gotom Anganty...
where Gautom is very possibly a miswriting for Grytingom — the Greutungi, a
Gothic tribe. See C.P.B. i, 349.
19. Becca. The Bikki of Northern story. See Introduction, pp. 19, 20, 33.
Mentioned again, 1. 115, as one of Eormanric's retainers. The facts that he
is here spoken of as prince of an independent tribe, that in Saxo he is son of
a Livonian king, and that the Atlakvi\>a mentions the "warriors of Bikki"
Bikka greppa (see C.P.B. i, 57, Symons u. Gering, i, 426-7), seem to indicate
that in the original story Becca was the prince of a tributary race, not a
Gothic servant of Ermanaric.
Baningum. Miillenhoff (Z.f.d.A. xi, 276-7) suggests that this is a fictitious
name " the sons of the slayers," <f>ovd5<u — a fit tribe for the faithless Becca to
rule. This would necessitate a : but on metrical grounds it is better to read a.
The name means then, presumably, " the righteous, hospitable " (Icel. beinn,
see Much, 126), a name which might have been assumed by a real people.
This interpretation seems preferable to the " Beininge " of Holthausen [of
course beinn may be connected with O.E. ban, Germ, bein, through the meaning
" the straight (i.e. shin-) bones "]. The probability that the Banings were a
real people is increased by an apparent reference to them in the Origo gentix
Langobardorum where the land of the Bains is linked with that of the
Burgundians, just as here Banings and Burgundians are classed together.
But this does not enable us to identify the Banings with any known tribe.
Schiitte (Oldsagn, 61, 106) suggests the Sarmatians. But evidence to prove
this is lacking. The Sarmatians had, Schiitte points out, a king Beuka
(Jordanes, LIV).
Burgendum. Originally an East Germanic tribe, and, in the second and
192 Widsith
20 Casere weold Creacum ond Gaelic Finnum,
third centuries, neighbours of the Goths. See Ptolemy, n, 11, 8-10 (where they
are mentioned as dwelling north of the Lygii: i.e. in the modern Fosen) and
perhaps in, 5, 8 (where if the Bnrgundians are to be identified with the
$povyoi{v]5lw}>es they must have extended east of the Vistula). From Jordanes
(xvn) it is clear that Goths, Gepidae and Burgundians were neighbours about the
middle of the third century, when Fastida was encouraged by his victory over
the Burgundians to attack the Goths under Ostrogotha. (Nam Burgundzones
pene usque ad internicionem delevit.) After this the Burgundians moved west,
and by the end of the third century were threatening Gaul. (Cum omnes
barbarce nationes excidium universes Galliae minarentur, neque solum Burgun-
diones et Alemanni, sed et Heruli et Chabones in has provincias irruissent —
Mamertinus, Panegyr. i, 5; 289-290 A.D. ) For the later history of the Bur
gundians see Introduction, p. 63. This mention of the Burgundians along
with the Eastern folk seems accordingly a reminiscence of a state of things
which had passed away before the fourth century (see H. Derichsweiler,
Geschichte der Burgunden, 1863 ; Bremer in Pauls Grdr. (2) in, 824), and hence
an indication of a very early date for at least this section of the poem. We
cannot be quite certain of this, however, for a portion of the Burgundian nation
may have remained in the East. Burgundians are reckoned by Sidonius
Apollinaris among the host of Attila invading Gaul in 451 (Cam. vn, 322), and
it has been supposed that these were "a portion of the tribe who had lingered
in their old homes by the Vistula " (Hodgkin's Italy, u, 107). Whether there
were any Burgundians by the Vistula so late as this may be doubted. Attila's
Burgundians may well have been a portion of the Bhenish people, which may
have been incorporated by the victorious Huns, after the Burgundian defeat in
437. The main body of the Burgundian survivors was, of course, in the
opposite camp (Jordanes, ed. Mommsen, xxxvi).
Gifica. See Introduction, p. 64.
20. Casere. The emperor of the East. The Kaiser's name was so current
in early Germanic times, that, at last, as the son of Woden, he comes to stand
at the head of the East Anglian genealogy. (See Florence of Worcester,
ed. Thorpe, i, 249.) He occupies this position in that genealogy as given
in MS. C.C.C.C. 183, and in MS. Cotton Vespasian B 6, where we should read
caser, not (as Sweet, Oldest English Texts, 171) care. The word casere may
well have been borrowed very early : possibly through Anglian mercenaries in
Roman pay (cf. Hoops, Waldbdume, 569). For Casere see also note in M.L.E.
iv, 1909, p. 508.
Creacum. The Greeks. Paul (P.B.B. i, 197) has argued that the C, which
is always found in this word both in the East and West Germanic dialects, and
which was only driven out later by the influence of the Latin Graecus, is due to
the fact that the last stage of the Germanic sound shifting was not completed
when the word was borrowed from the Latin, and that hence the voiced stop
became voiceless along with the other similar sounds. This would make the
word a very early loan indeed. A more probable theory is that of Kossinna
(Weinholdfestschrift, Strassburg, 1896, p. 40) and Kluge (Pauls Grdr. d) i, 325)
who suppose that, at the time the word was borrowed, the voiced stop g had
not come again into use at the beginning of words, and that thus the voiceless
stop k was substituted, as the nearest approximation.
For the spelling Creacum for Crecum, here and in 1. 76, cf. Holthausen,
165. See also Introduction, p. 166.
Ccelic. Heinzel (Hervararsaga in W.S.B. cxiv (1886) p. 507) suggests that
this is a corrupt form of Kalew. It is certainly the case, as Heinzel points
out, that stranger corruptions are found in the Anglo-Saxon and Old Norse
genealogies. The MS. has C$lic ( = Gaelic).
21. Hagena. See Introduction, p. 100, etc.
Holm-Rygum. An emendation of Grimm (G.D.S. 469 ; following a suggestion
of Lappenberg, 175) for the MS. Holmrycum. Grimm's conjecture is certain
(cf. 1. 69) : it has been followed by later editors. Ettm., Thorpe, Ebeling,
Klipstein, and Grein, however, follow Leo, Holmricum, or MS. Holmrycum.
Text (II 20, 21) 193
ax Hagena Holm-Ry^rum ond Heoden Gloramum.
The Bugian name, like the Gothic, is found both in Scandinavia and in
Eastern Germany (Pauls Grdr. (2) in, 818; Much in P.B.B. xvn, 184). The
Scandinavian Rygir, Holmrygir (Island-Eugians) dwelt in Bogaland, on the
banks of the southernmost islands and fiords of Norway. But not even the
Norwegian versions place Hogni in Norway. Saxo puts him in Jutland, and
Snorri imagined him as dwelling well to the south of Norway. The Holmryge
are therefore more probably the Rugii of Tacitus, a people dwelling on the
Baltic coast not far from the Goths (protinus delude ab Oceano Rugii; Ger-
mania, XLIII). That these were also known as ' ' Island-Bugians " is clear from
Jordanes, according to whom the Goths, on leaving Scanzia, reach the main
land unde max promoventes ad sedes Ulmerugorum, qui tune Oceani ripas
insidebant, castra metati sunt, eosque commisso proelio propriis sedibus pepu-
lerunt, eorumque vicinos Vandalos iam tune subiugantes suis aplicavere victoriis
(ed. Mommsen, iv). The Island-Bugians must have dwelt on the islands and
peninsulas at the mouth of the Vistula (Miillenhoff in Mommsen's Jordanes)
and perhaps also of the Oder (Zeuss, 484), whence the modern Biigen. It is to
be noted that, according to Saxo, the last fight of Hoginus and Hithinus took
place on the island of Hiddensoe (Hythini insula) off the coast of Biigen.
Later the Bugians moved south, and in the fifth century occupied the
present Austria. The story of Hagena and Heoden, essentially one of sea
roving, could accordingly no longer be associated with them. Hence the
confusion of the later versions : Kudrun places Hagen in Ireland : the Sgrla
\>dttr and Saxo in Denmark.
That Widsith still connects the story with a tribe which, by the fifth
century, had ceased to be maritime, is another proof of early date: but the
possibility that the reference is to the Norwegian Holmrygir must not be left
quite out of sight. Compare Koegel, Ltg. i, 1, 169 ; Miillenhoff in Z.f.d.A. xxx,
230; Pauls Grdr.(2) in, 826-7; Panzer, H.G. 436-7. Boer (Z.f.d.Ph. xxxvm,
47) would make the Danish islands the original home of the " Island-Bugians."
But there is no evidence for this. Bugge (H.D. 314-5) sees a connection
between the Holmryge and the Bogheimr of Helga kvffia Hjyrvarftssonar, 43.
This again seems unlikely.
Heoden. MS. Henden. The emendation Heoden (which is certainly right,
although Schaldemose, Ettmiiller^), Thorpe, Ebeling, Klipstein, and Grein
retained the MS. reading) was suggested by J. Grimm (Z.f.d.A. n, 2 ; cf. G.D.S.
470). Moller suggests an intermediate Headen : so, independently, Anscombe
in Anglia, xxxiv, 527 : in view of 11. 70, 111, it is quite likely that Heoden was
written Headen : and ea would then easily be miswritten en.
Glommas. Lappenberg (175), Miillenhoff, Nordalb. Stud, i, 151, Grimm,
G.D.S. 752, Thorpe, Grein (Sprachschatz), and Holt, suggest a connection with
Norway (the river Glommen). It is in favour of this conjecture that Saxo
places Hithinus in Norway, as apparently does the catalogue of the battle of
Bravalla (Hythin gracilis, in Saxo, vin, p. 258). See also Introduction,
p. 109, note.
Conybeare would make the Glommas " a Sorabic tribe," his evidence being
apparently Ditmar v. Walbeck, who speaks of " provintiam quam nos Teutonice
Deleminci vocamus, Sclavi autem Glomaci appellant " (Thietmari Chronic&n, i,
3, recog. F. Kurze, Hannoverse, 1889). Glomazi is the modern Lommatzsch
near Dresden. The ancient province coincides approximately with the later
Erzgebirgischer Kreis of the electorate of Saxony. For the exact limits,
cf. Kurze's note to Ditmar, and Bottger's Diocesan- u. Gaugrenzen Norddfutsch-
lands, iv, 224, etc. That there is any connection between Glomazi and
the Glommas is, however, in the highest degree unlikely. Still more so is the
suggestion of Koegel (Ltg. i, 1, 169) that, as Heoden's folk, the Glommas should
be sought near the mouth of the Schelde. This is an application to Widsith of
features belonging to a much later version of the story.
The only evidence then, though that is insufficient, points to Norway as the
home of the Glommas.
c. 13
Widsith
22 Witta weold Swsefum, Wada Hselsingum,
22. Witta occurs in the genealogy of Hengest as given by Bede, H.E. i, 15 ;
and following Bede, by the Laud MS. of the A. S. Chronicle : Hengest and Horsa,
\<zt wcer&n Wihtgilses suna. Wihtgils wees Witting, Witta Wecting, Wecta
Wodning. The same genealogy is also given in the Historia Brittonum, § 31,
cf. Grimm, D.M. Anhang, iv, v. A not very probable conjecture by Fahlbeck
as to the origin of this genealogy will be found in A.T.f.S. vm, 48. The name
Witta is found associated with that of Watte (? = Wade) in a tradition of
Northern Schleswig, recorded by Miillenhoff. Vitte and Vatte are two dwarfs
dwelling in neighbouring hills : a farmer passing one hill at night is told to
" greet Vitte, Vatte is dead." See Mulleuhoff, Sagen, Mfirchen und Lieder der
HerzogtMlmer Schleswig-Hohtein u. Lauenburg (1848), No. 400, p. 242. In
England the name is found in Wittanham in Berkshire (see Binz in P.B.B. xx,
221).
Swcefum. Swabians, Suevi. The name is used, broadly, to cover a large
number of tribes in central Germany : Nunc de Suevis dicendum est, quorum
non una ut Chattorum Tencterorumve gens ; majorem enim Germanics partem
obtinent propriis adhuc nationibus nominibusque discreti quamquam in commune
Suevi vocentur. (Germania, xxxvin.) Schwabsted in Schleswig, compared
with the South German Schwaben, perhaps bears witness to this wide diffusion.
But Tacitus probably exaggerates the extent of the Suevic name, making
it cover the Angles, and many tribes of Eastern and Northern Germany who
were very probably not of Suevic stock. That the Suevi extended as far as
the coast is probable from a passage in the Agricola (xxvni) where some
shipwrecked men fall first into the hands of the Suevi, then of the Frisii
(primum a Suevis, mox a Frisiis intercepti) ; see Chadwick (203). Amongst this
large family of Suevic tribes there was, however, one tribe to which the name
peculiarly belonged : the Suebi of Caesar, dwelling on the Main. These, after
many wanderings, mingled with the Alamanni, themselves Suevi in the broader
sense of the word, and to this mingled stock the name Swabian has in later
times been specially applied. See Bremer in Pauls Grdr. (2) m, 918-38. The
name had therefore two significations, one broader, the other more particular.
Below, where the Swcefe are linked with the Angles (11. 44, 61), the reference is
apparently to the northernmost section of the great people, dwelling near the
modern Schwabsted (Much, 28, 102) : and here the context favours a similar
explanation. Whereas in 1. 44 the Swsefe are apparently synonymous with the
Myrgings, it should be noted that here the two peoples are clearly distinguished.
Wada, see Introduction, p. 95, etc.
Halsingum. Miillenhoff (Z.f.d.A. vi, 65; cf. xi, 278) regards this as a
feigned name, connected with hals, the neck, or prow of a ship (flota fami-
heals, Beowulf, 218) chosen as appropriate to the people of a hero, like Wada,
connected with the sea. More probably the Hselsingas are a real people, whose
colonists have left their name in many places on the Baltic coast : Helsingor
(Elsinore) in Zealand ; Helsingborg on the adjacent coast of Sweden ; Helsing-
land, a province of Sweden ; Helsingfors, on the opposite coast of Finland. In
spite of this wide diffusion, there is considerable agreement among philologists
in placing the original home of the Hselsingas on the XdXoixros 7roTOyoi6s of
Ptolemy, but great disagreement in identifying this with any modern river.
Zeuss (150), Seelmann, who argues the question at length (43), Weiland (27)
and Chadwick (204) suggest the Trave; Moller (V.E. 28) the Eider or Halerau ;
Miillenhoff (Nordalb. Stud, i, 115) the Eider ; Much (P.B.B. xvn, 186) the
Warnow. Cf. also Moller, A.f.d.A. xxn, 155.
In any of these cases the Hselsingas would be neighbours of the Angles, and
probably of "Anglo-Frisian" stock. (Kogel, Ltg. i, 1, 155-6.) If, as is
probable (cf. Panzer, H.G. 436), the Middle High German Wate von den
Stilrmen means Wade "of Stormarn " this would be a strong argument for
placing Wade's people, the Healsingas, in that district, i.e. on the Trave. In any
case Wade and his people are to be put on the Baltic coast : and with this the
reference in the Thidreks saga agrees (c. 57, Vafte rise er a Siolande).
Text (II 22-24) 195
13 Meaca Myrgingum, Mearchealf Hundingum.
peodric weold Froncum, pyle Rondingura,
23. Meaca. Mullenhoff (Nordalb. Stud, i, 152) followed by Bieger, conjectured
Meara, the eponymous form from which Myrging might be supposed to be
derived. He abandoned the conjecture (Z.f.d.A. xi, 277) but returned to it
(Beovulf, 99). In Z.f.d.A. xi, 277 Mullenhoff would connect Meaca with O.E.
gemaca, Icel. maki, companion, and would interpret Mearchealf as "colleague."
He apparently regarded the two as forming a heroic pair, the Theseus and
Pirithous of early German story.
Mearchealf. Ettmullerd) suggested Mearcvulf (notes, p. 15) but returned
to Mearcealf in his second edition. Mearcvulf was also suggested in a note by
Bieger, who however retained Mearchealf in his text, following Miillenhoff's
explanation (see Meaca, above). If we read Mearcwulf we must interpret as
the Marculf of Salomo and Saturn. Sophus Bugge (H.D. 171) has independently
made the same suggestion as Ettmiiller, pointing out, what is at any rate an
odd coincidence, that Marculf is spelt Marcholfus by Notker. Marculf 's son
and successor in Middle English verse is Hendyng. It would be rash to argue
any connection between this name and the Hundingas whom Mearchealf rules
(cf. Binz, P.B.B. xx, 222).
Hundingum. King Hunding is mentioned in the prose introduction to the
song of Helgi Hundingsbani, Hundingr het rikr konungr, vfy hann er Hundland
kent (Symons, Edda, i, 272). Tribal and kingly names tend to interchange :
cf. Hoeing, below, and Scylding. The Hundings are apparently to be placed on
the south coast of the Baltic See Mullenhoff in Nordalb. Stud, i, 153, Z.f.d.A.
xi, 278, xxni, 128, 170 ; Jiriczek, 292 ; Bugge, H.D. 91, 169-70; P.B.B. xxxix,
479. Hund may be either from hund "dog," or hund "hundred"; in the
former case we may compare the cynocephali of Paulus (i, 11), see also
Anscombe in Eriu, iv, 87 ; for the latter compare Tacitus' mention of the
hundred pagi of the Semiiones (German in, xxxix). For another interpreta
tion, see Chadwick, 299-300.
24. peodric. See Introduction, p. 112. Moller (V.E. 18) supposes Theodric
to be an interpolation from some poem dealing with the victory won by Theodric
over Hygelac. There is little to be said in favour of this conjecture, which
seems to spring from the very common habit of exaggerating the importance of
those remnants of the old epics which happen to have come down to us. The
rout of Hygelac was only one of Theodric's victories, less important and
less personal than many others.
Froncum. If the name of the Franks means the " spearmen," as argued by
Kluge (Etymol. Worterbuch) and, with some doubt, by Erdmann (Die Angeln,
80), then we might take the "Bondingas " as an epic fiction echoing the name.
"Theodric ruled the Spearmen: Thyle (his retainer) the Shieldmen."
But it seems unlikely that the Franks derived their name from the franca,
for two reasons. (1) As has been frequently noted, the root frank = spear is not
found in continental German, but only in English and Scandinavian dialects.
(2) Equal attention has not been given to the argument that the franca was
used certainly as much, possibly more, by other nations. The characteristic
Frankish weapon was the hurling axe, frequently referred to by Procopius (see
Keller, M.L., Anglo-Saxon Weapon Names, Heidelberg, 1906, pp. 18-31, 133-4).
According to Agathias (n, 5) the Franks used a peculiar barbed spear, the anyo
(cf. our angle) similar to the " cruelly hooked boar spear " which in Beowulf is
used to attack sea-monsters. Agathias gives a detailed description of the ango :
it was a species of harpoon, and could not be withdrawn from a wound. Some
good illustrations of these weapons, recovered from the graves of Bipuarian
Franks, will be found in Archteologia, xxxvi, 78-84 (cf. also xxxv, 48-54).
There seems to be no connection between the ango and the O.E./ranca, which
was a gentlemanly weapon :
he let his francan wadan
tzs hysses hals.
13—2
196 Widsith
25 Breoca Brondinguwi, Billing Wernum.
According to Procopius, the Franks in A.D. 539 were remarkable for not carry
ing spears: lirvtas fiev 6\lyovs rivas dfi<f>l rbv riyotj/jievov ^o^Tes, ot dr) Kal /J.6voi
S6pa.ro, H<pepoi>. ol Xotirol S£ vefbi airavres oCre r6£a afire S6pa.ro. £xo»res, dXX& £/0os
re KO.I do-vLda <ptpwv g/caerros Kal ir£\CKW &/a. (Bell. Gott. II, 25.) It is very
doubtful, therefore, if the word Frank conveyed the idea of " spearman." The
rival etymology " the Freemen" suggested by Grimm, G.D.S. 512, etc. is also
open to grave doubt: it has been supported by Bremer in Pauls Grdr.(2)
HI, 878. (But cf. P.B.B. xxv, 223, etc.)
pyle. See Introduction, p. 115.
Rondingum. Perhaps an epic fiction, the " shield men " (a name appropriate
to the retainers of Thyle, Theodric's follower) as Mullenhoff suggested. No
historic tribe of this name is known : it certainly means " men with shields," cf.
Icel. hyrningr, one carrying horns. (See Much in P.B.B. xvu, 120 ; Binz in
P.B.B. xx, 148; Miillenhoff in Z.f.d.A. xi, 281; Chadwick, 284.) Leo (77),
Conybeare (12), Lappenberg (178) and Brandl (Pauls Grdr. 966) suppose the
Bondings to be a real people, and their name to be preserved in the Beudigni
of Tacitus : a somewhat unlikely conjecture. For these Beudigni see Much in
P.B.B. xvn, 192.
25. Breoca Brondingum. Breca also in Beowulf rules over the Brandings.
See Introduction, p. 110.
Billing. This name is best known in connection with the famous house
which ruled in Saxony during the latter part of the tenth and the whole of the
eleventh century. See Prutz, Staatengeschichte des Abendlandes im Mittelalter,
1885, i, 189, 400 ; E. Steindorff, De ducatus qui Billingorum dicitur, origine,
Berolini, 1863 ; E. Wintzer, De Billungorum intra Saxoniam ducatu, Bonn®,
1869. The name is of frequent occurrence in England, Lappenberg's statement
(Geschichte, i, 214) notwithstanding, and has left its mark in names like
Billingborough, Billinge, Billingham, Billinghay, Billingshurst, Billingsgate,
Billington. Of the original Billing nothing is known, but we may assume,
from his connection here with the Werne or Waernas, and the subsequent
appearance of the name in England and Saxony, that he was essentially a hero
of the coast tribes. See also Grimm, D.M. (219 : trans. Stallybrass, i, 373-4) :
Kemble, Saxons, i, 458 (Appendix A). For name (bill, sword) cf. Ronding above.
Billing's daughter is the beloved of Odin : Hdvajndl, 96.
Wernum. The Varini of Tacitus, a people closely connected with the
Angles, and belonging to the group of Nerthus-worshipping tribes. We gather
from the order of Tacitus' enumeration that the Varini probably dwelt North of
the Angles, and their name may perhaps be preserved in Warnces (i.e. Warna
nses, promontorium Varinorum, now corrupted to Warnitz, a promontory in
the N.E. of Sundewitt). See Mullenhoff in Nordalb. Stud, i, 129. Procopius,
in the middle of the sixth century, makes the OOapvoi spread as far as the
Bhine, which separates them from the Franks : dnJKovfft 6t OX/M re fs &iceavbi> rbv
apKTtpov Kal iroTa.iJ.bv 'Pfjvov Sffwep atfrotfs re dioplfei Kal &pdyyovs. (Bell. Gott. iv,
20, 1833; a strange story follows of a war between the Angles of Brittia and
the Warm.) Another reference (11, 15) marks the Warni as dwelling more
where we should expect them : south of the Danes. The Varini can hardly
have occupied these wide stretches of country continuously, as Hodgkin (Italy,
in, 355) seems to suppose. They must have been scattered, and settled in
isolated groups. Hence the difficulty of localizing the Werini of the Lex, and
the Guarni addressed by Theodoric. (See Appendix E : The original home of
the Angli and Varini.) That a contingent also accompanied the Angles to
England appears likely from the names Wernanbroc, Wernanford, Warnanhyll,
Wernanwyl. (See Seelmann, Die Ortsnamenendung -leben in the J.d.N.f.n.S.
xn, 1886, p. 23 : for other place names, many very doubtful, see Haigh
(113) ; certainly not Warwick as suggested by Miiller, Lex Salica, 123 ;
corrected by Mullenhoff, Nordalb. Stud, i, 133. See also Zeuss, 360 ; Mullen
hoff, D.A. iv, 465; Much in P.B.B. xvn, 204; Bremer in Pauls Grdr.(2)
in, 851.)
Text (II 25-29) 197
16 Oswine weold Eowum ond Ytum Gefwulf,
Fin Folcwalding Fresna cynne.
Sigehere lengest Sse-Denum weold,
Hnaef Hocingum, Helm Wulfingum,
26. Oswine. Unknown, unless he is to be identified with the Oslaf mentioned
in the Fhmesburh episode in Beowulf. See Much in Herrigs Archiv, cvm, 407
and cf. Grimm, G.D.S. 472, and Miillenhoff in Z.f.d.A. xi, 281.
Eowum. Allowing for the interchange of eo and ea (Beowa and Beawa,
Z.f.d.A. vn, 411) and assuming a weak nominative Eawan, we have the Aviones
of Tacitus (see Miillenhoff, Nordalb. Stud, i, 156; Grimm, G.D.S. 472 ; Koegel,
Ltg. i, 1, 155). The Aviones belong to the group of Nerthus-worshipping
people. Much (P.B.B. xvii, 195) would locate them on the islands of the
Cattegat, interpreting the name as " Islanders." Bremer (Z.f.d.Ph. xxv, 129)
suggested that the name was preserved in the traditions of the North Frisian
islands, as recorded by Hansen. In the muster of warriors in the Sylt story we
have "De Uuwen kam fan Uasten," the Uuwen came from the east; aud
Bremer derived Uuwen from a primitive Auwaniz, whence Aviones, Eawan : see
C. P. Hansen, Uald Sold'ring Tialen, 1858, p. 22, and, dissenting fromBremer's
suggestion, Moller in A.f.d.A. xxn, 158. Subsequently Much (2 101, and in
Hoops Reallexikon, i, 146), whilst adhering to the explanation of Eowan,
Aviones as "Islanders," would place them in the North Frisian islands of the
North Sea rather than in the Baltic.
It has also been proposed to connect the Eowan with Oeland (called Eowland
in Alfred's Orosius, ed. Sweet, p. 20). But this would take us into quite
another part of the Baltic. We need an identification of the Eowan which
leaves them neighbours of the Waernas and of the Yte. This is supplied by the
connection with the Aviones, but not by the connection with Eowland.
Ytum. See Appendix, Note D : The Jutes. Gefwulf is unknown.
27. Fin Folcwalding. See Introduction, p. 67.
Fresna cynne. See Introduction, p. 67, and compare Zeuss, 136 etc. ; Grimm,
G.D.S. 668-81 ; Erdmann, 83-86 ; Siebs in Pauls Grdr. (2) i, 1153 ; Bremer in
Pauls Grdr. (2) in, 844-849. Cf. also note to 1. 68 below.
For the structure of the two lines, 26, 27, see Neckel, pp. 5, 6, 47, 480.
28. Sigehere. Undoubtedly identical with the O.N. Sigarr (Heinzel in
A.f.d.A. xvi, 272). Saxo, Bk vn, tells how Sygarus, king of Denmark, caused
Hagbarthus, the lover of his daughter Sygne, to be slain, and how Sygne shared
her lover's death (ed. Holder, pp. 230-237). The story is a particularly fine
one. It must have been popular throughout the Scandinavian countries :
references to it are frequent, especially in the form of place names (Heinzel).
The tale is also extant in a Danish ballad, of which there are many versions.
See Danmarks Gamle Folkeviser, udgivne af Svend Grundtvig. Kjtfbenhavn,
1853 etc. i, 1, 258-317. But in England it seems to have been little known,
and the name Sigehere is rare. (Binz in P.B.B. xx, 169 ; cf. Miillenhoff,
Beovulf, 52.)
lengest. Sygarus is styled senilis by Saxo (ed. Holder, 237). I understand
lengest here, and in 1. 45, below, to mean "for a very long time." Cf. the use
of swi~$ost for "very greatly" in Alfred's Boethius (p. 37, ed. Sedgefield, 1899).
29. Hncef Hocingum. Strictly Hnaef is himself the son of Hoc : but the
family name is extended to his people, just as in Beowulf the Danes are called
Scyldingas, from the ancestor of their kings. Hoeing is perhaps found for
the hero Hoc, the father of Hnaef, in the genealogy of Hildegard the wife of
Charles the Great, whose grandfather and greatgrandfather bore the names
of Nebi [? = Hnasf] and Huochingus (see Miillenhoff in Z.f.d.A. xi, 282). Hilde
gard was of Alamannic race, and we have therefore in the juxtaposition of
these two names an indication that the story was known in High Germany in
the seventh century. The names of the heroes of this cycle are found fairly
frequently in England, but more particularly that of Hoc (see Binz in P.B.B.
xx, 179-186).
198 Widstith
30 Wald Woingum, Wod pyringuw,
For the use of Hoeing both as a family name, and as equivalent to that of
the ancestor Hoc, cf. Vylsungr, which is applied not only to members of
the family, but also to the eponymous ancestor, who, in Beowulf, is called,
more correctly, Wcels.
Helm. Perhaps the imaginary name of a typical king: cf. the common
expression helm Scyldinga, Scylfinga. But it is no mere fiction of our poet, for
that there was a hero Helm, though his story has been lost, is likely from the
mention of a Helmes treow (Cart. Sax. ed. Birch, in, 1213) whilst a Helmes
welle is mentioned several times in Domesday Book. The Helming family
— presumably that of which Helm was the eponymous ancestor — is referred to
in Beowulf (620), and place names derived from Helming are fairly common in
England (see Haigh, 63 ; Binz in P.B.B. xx, 177-8). Continental parallels are
not so easy to find : but there seems reason to suppose, with Ettmiiller d) 17,
some connection with the Helmschrot or Helmschart of German story (Dietrichs
Flucht passim). At any rate Helmschrot, Helmschart and Helmnot all belong
to the Wiilfinge (cf. Matthaei in Z.f.d.A. XLVI, 55-6).
Wulfmgum. The name occurs in Beowulf. Ecgtheow having incurred
a blood feud mid Wilfingum, Hrothgar pays the fine and sends ofer wceteres
hrycg ealde madman. The phrase seems to imply a sea-journey of some
distance. As the Wiilfinge, the faithful retainers of Dietrich von Bern,
play an important part in the High German versions of Gothic legend, it
is likely that the original Wylftngas should be thought of as neighbours of the
Goths, dwelling on the Baltic coast. See MuUenhoff, Beovulf, 14, 90 ; Z.f.d.A.
xi, 282 ; xxxm, 170 ; Jiriczek (2), 292. That the Wylfmgas of the O.E. heroic
verse are the same as the Wiilfinge of the Heldenbuch is the more likely
because in both cases members of the family form their names by compounds
in heafto- ; e.g. Heaftolaf, Hadubrant (Binz in P.B.B. xx, 216) ; see also the
preceding note on Helm. Traces of the name are to be found in England:
Wolfinges law, Wylfingaford. The Ylfingar occur in Scandinavian poetry:
Helgi Hundingsbane is called "the son of the Wolfings" Ylfinga raft. Cf.
Bugge, H.D. 84-87 ; 164-166 ; 313-314 ; see also Symons in P.B.B. iv, 176 ;
and, for some further notes on the Wolfings, and on Helm, Sarrazin in Engl.
Stud, xxm, 228 : Heinzel, Hervararsaga in W.S.B. cxiv, 510.
30. Wald. Wald is regarded by Miillenhoff as an epic fiction, the typical
name of a tyrant (Z.f.d.A. xi, 283).
Woingum. In Nordalb. Stud, i, 158, Miillenhoff suggested a connection
between the Woingas [primitive form *Wanh-~\ and the island of Wangeroge,
and the district Wangerland, the older Wangia or Wanga. (See the Chronicon
Moissiacense and the Vita S. Willehadi in Pertz (fol.) SS. n, 257 and 383.)
They would therefore be the inhabitants of the coast of East Friesland and
Oldenburg. In his later commentary (Z.f.d.A. xi, 283) Miillenhoff proposed to
take the name of the Woingas as an epic fiction, as well as that of their king: a
typical name for an evil people, from woh "crooked," "evil." This last
explanation seems less satisfactory. Miillenhoff's further development of it
(Z.f.d.A. xxni, 124), in which he attempts to connect Wald and his Woings
with the family of the tyrant Siggeir of the Vglsunga Saga, wants evidence to
support it.
Wod. Here again Miillenhoff suggests that the name is fictitious and
chosen with reference to the Thuringians, over whom Wod (i.e. mad) rules:
(there was a popular etymology which connected Thuringian with thor " fool ").
But see Grimm, G.D.S. 597.
For the Wod of Germanic mythology (O.N. OSr, the husband of Freyja,
Hyndloljofi, Vglospd), see Golther, 283 etc. We cannot say whether Wod of
the Thuringians has any connection with him.
pyringum. The pyringas may be the Thoringi of Gregory of Tours (Hist.
Franc, n, 9) who dwelt in the present Zeeland and North Brabant, in the neigh
bourhood of Dordrecht, at the mouths of the Bhine and Maas. These Holier
(V.E. 16) identifies with the Turii or Sturii who are mentioned by Pliny as
occupying the same territory. (Nat. Hist, iv, cap. 15 [29].) These Thuringians
Text (It. 30, 31) 199
31 SseferS Sycgum, Sweom Ongenctyeow,
were apparently a Frankish people, who were subsequently pressed South
and East by the Saxons and Frisians. It was Hermann Miiller (Der Lex Salica
u. der Lex Angliorum et Werinorum Alter u. Heimat, Wiirzburg, 1840, p. 122
etc.) who first pointed out the existence of these Thoringi, and the necessity
of distinguishing them from the Thuringians proper ; cf. Waitz, Das alte Recht
der Salischen Franken, Kiel 1846 ; Grimm, G.D.S. 601 ; Moller (A.f.d.A. xxn,
152) ; Erdmann (28). As we are dealing here with North Sea peoples and
heroes, it is probably these Ehenish Thuringians who are meant (cf. Mullenhoff
in Nordalb. Stud, i, 159). It is possible, however, that here as in Alfred's
Orosius (ed. Sweet, 1883, p. 16) pyringas means the main stock of the Thurin
gians, identical with the Hermun-Duri of classical times, whose name is in the
fifth century replaced by that of Thuringian : (the Thuringians, however, cover
a wider area, from the Harz mountains to the Danube). The Thuringian king
dom was overthrown by Theodoric, king of the Franks (cf. 1. 24) in 531-6. See
Zeuss (353 etc.) ; Bremer in Pauls Grdr. (2) in, 938 etc. ; Grimm, G.D.S.
596-607 ; Seelmann (Nordthttringen) in the J.d.V.f.n.S. 1886, 12.
31. Scefer%. Identical with the SigeferS, Secgena lead of Fin. 1. 42. For the
confusion of See and Sige, cf. Scsberht, king of the East Saxons, who appears as
Saberchtus or Saeberchtus in the text of Bede, but as Sigberchtus (at any rate in
certain MSS.) in the Chronological Summary (cf. Moller, V.E. 87). The name
SigeferS is common in England, and so, to a less degree, is S<zferft (Binz in
P.B.B. xx, 184-5). It has been argued by Uhland that this Sigeferth or Saeferth
is to be identified with Sigfried (Sigurd) the Volsung (Germania, u, 357-63 ;
Schriften, vni, 497). This view has been more recently supported by Golther
(Germania, xxxni, 474-5) and apparently for a time by Mullenhoff (see the
next note). It seems exceedingly improbable, though it is difficult to speak
with certainty, in the absence of sufficient evidence as to the Sigfried story
in England.
Sycgum. Koegel (Ltg. i, 1, 164) and Moller (V.E. 86) connect the Secgan
with the Saxons " for Secga like Sahso signifies swordsman " (Koegel). Secgan
and Seaxan would then be related as secg, "sword" (Beowulf, 684) and seax
"knife," from Prim. Germ, sag-, sahs- (cf. Lat. secare). These Secgan Moller
supposes to have dwelt between the Elbe and the Eider, and to be identical
with the tribe who colonized Essex. The evidence for this, however, is not con
vincing, and turns chiefly upon the occurrence of the names Gesecg and Antsecg
in the Essex genealogy, together with Seaxnete, the son of Woden, with whom
the genealogy begins. (See genealogies in Florence of Worcester, ed. Thorpe for
English Historical Society, 1848, i, 250 ; and Henry of Huntingdon, Historice
Anglorum, u, § 19, ed. Arnold, Bolls Series, p. 49.) Mullenhoff (Nordalb. Stud.
i, 164) had drawn attention to Gesecg and Antsecg : he suggested an identifi
cation of the Secgan with the Beudigni of Tacitus. All this is exceedingly
hypothetical, but it seems likely, from the context in which they are mentioned
here, in 1. 62, and in the Finnesburh fragment, that the Secgan were a coast
tribe. Uhland, having identified Sigeferth with Sigfried, had to identify the
Secgan with his people, deriving the name from Sigi, who in the Vglsunga Saga
stands at the head of the family [precisely as the Danes are called Scyldings
because Scyld is the first of their royal house]. This identification was rejected
by Mullenhoff in 1859 (Z.f.d.A. xi, 283) but accepted by him in 1877 (Z.f.d.A.
xxni, 117-124), although with a full recognition of the etymological difficulty
of bringing Secgan and Sigi together (155-7). Finally, however, Mullenhoff
recognized this identification as unlikely (Beovulf, 97) whilst at the same time
rejecting Moller's explanation.
Lappenberg (177) suggests a connection with the SiyotiXuves (recorded by
Ptolemy in the Cimbric Chersonese, next to the Saxons). Geographically
this is satisfactory, but linguistically the gap is too great, as corresponding to
Secgan we should expect Sagiones.
For Secgum : Sycgum, see Moller, V.E. 153-4.
Sweom. The Swedes. The name however at this date embraces only a small
section of the inhabitants of modern Sweden, which was held by Finns in the
200 Widsitk
yi Sceafthere Ymbrum, Sceafa Longbeardura,
North, Norwegians in the West, Geatas and Danes in the South. See note to
p. 72. Moller (V.E. 15) considers that the name of the Swedes is substituted
here for that of some other tribe beginning with S — Sweordwerum, Sweardum,
or Seaxum — on the ground that the transition from tribes at the mouth of
the Elbe to the Swedes is too sudden. But it does not follow, because the
names in Widsith are usually grouped geographically, that they must always
be so.
Ongend\>eow. The king of Sweden, frequently mentioned in Beowulf (2476,
etc. : 2923, etc.). The name Ongen]>eow is unknown in England in historical
times, nor is any tradition of him retained, beyond the references in Beowulf
and in Widsith. A reminiscence of the name is to be found in the fact that the
son of Offa I, called AngeVpeow in the Chronicle, is called Ongen in the Historia
Brittonum (in M.G.H. p. 203). Moller (V.E. 19) regards the name here as an
interpolation — but the reasons brought forward are insufficient. Ettmuller (t) (2)
and Grein read Ongerfyeow, to bring the name into line with the forms used in
Beowulf.
For the story of Ongenthew, which is clearly in great part historic, see
H. Weyhe, Konig Ongentheows Fall, in Engl. Stud, xxxix, 14-39 ; Gering, 115,
note, 118 ; Bugge in P.B.B. xn, 16.
32. Sceafthere. Possibly, like Helm, Scild, or Msc, a typical name for a
war-king (Miillenhoff in Z.f.d.A. xi, 283 ; cf. also vn, 414).
Ymbrum. Perhaps the inhabitants of the island Fehmarn and the adjoining
coast. Fehmarn is repeatedly called Ymbria, Imbra in Danish chronicles
and documents (see Langebek, passim), as also by Adam of Bremen (Pertz (fol.),
SS. vn, 373, where compare the note). This suggestion was originally made
by Lappenberg (p. 177), and has been again made independently by Siebs
(Eng. Fr. Sprache, 17). It is the best explanation of the name.
But Thorpe (Cod. Ex. 516) and, apparently independently, Miillenhoff
(Nordalb. Stud, i, 159 ; cf. also Z.f.d.A. xi, 283) would connect the Ymbre with
the district of Ammerland, west of the Weser, known in the llth century as
Ammeri or Ambria (see Staphorst, Historia Ecclesia Hamburgensis Diplomatica,
Hamburg, 1723, i, 425, and i, 603, where in a charter of ? 1199 we have comes
Elimer Ambria ; see also Ledebur, Die filnf Miinsterschen Gaue und die sieben
Seelande Frieslands, Berlin, 1836, especially map at end: and H. Bottger,
Diocesan u. Gaugrenzen Norddeutschlands, Halle, 1875-6, n, 161). Perhaps,
also, the name is preserved in Amrum, one of the North Frisian islands.
Cf. also Ettmiiller(i) 18; Much, 101.
It may be that in these two names Ambria, Amrum we have traces of
fragments of the famous tribe of the Ambrones, for whom see Lappenberg,
i, 101-2; Zeuss, 147-150; Miillenhoff, D.A. n, 114, 118, Beovulf, 95. The
Ambrones are mentioned by Plutarch as taking part in the expedition of the
Cimbri and Teutones (ed. Reiske, Lipsise, 1775, n, 830, Life of Marius, Teirrom
d£ Kal'An^puves tyalvovro ir\-f)dei T &ireipoi xa.1 dvffirp6ff(airoi ra ef&ri). They seem,
in fact, to have early got a bad name, and the word Ambrones, as is clear from
the many instances quoted by Miillenhoff, is often used simply as a synonym
for robbers.
There is some evidence, though far from conclusive, that a section of the
Ambrones took part in the colonization of England. Siebs (Eng. Fries. Sp. 15,
16) sees many traces of their name in ancient Wessex (Worcestershire and
Wiltshire) : and the close connection which many students have seen between
the North Frisian dialect of Amrum and Old English might also be urged
in support (cf. Moller in A.f.d.A. xxn, 157). The heavy evidence for a settle
ment of Ambrones in England is that the Historia Brittonum, describing the
baptism of the Deirans (more correctly the Bernicians, as Ten Brink has
pointed out, referring to Bede, H.E. n, 14), calls them Ambrones : et per quad-
raginta dies non cessavit baptizare omne genus Ambronum, where a 13th century
MS. (Corp. Christ. Camb. 139) records the gloss id est Ald-Saxonum. See
Nennius, ed. Stevenson, Eng. Hist. Soc. 1838, § 63 ; Moller, V.E. 89 ; Ten
Brink, Beowulf, 198-9.
Text (tt. 32, 33) 201
33 Htin Hsetwerum ond Holen Wrosnum.
Perhaps, however, the use of the term in the Historia Brittonum is not
ethnographical at all, but expresses the previous unregenerate state of the
converts : the gloss may then be the conjecture of some scribe who did not
understand this; or, as Chadwick suggests (56), Ambronum may be an error
for Umbronum, men of the Humber. (Cf. also Latham's Germania, 1851,
cvin.)
Mullenhoff's connection of the Ymbre with Ambria has been generally
followed (e.g. by Bieger in Z.f.d.A. xi, 202-4), but Lappenberg's identification
with Imbria is far more satisfactory. Whether there is a connection between
any or all of the names Ambria, Amrum, Imbria and the classical Ambrones
cannot be decided with certainty, but such connection is very probable. The
differences of vowel would presumably be due to ablaut (Much in Hoops
Beallexikon, 77).
Sceafa. Binz, following Moller, would read here Scyld, or Scefing (P.B.B. xx,
148). He continues " Sceafa ist iibrigens auch wegen der schwachen form sehr
verdachtig." But strong and weak forms of the proper noun interchange not
infrequently. Thus we have Sceldwea, Sceldwa in the Parker MS. of the
Chronicle (sub anna 855), in Asser, and in Florence of Worcester, as against
Scyld in Beowulf, Ethelwerd, etc. : Geat in the Parker and most MSB. of the
Chronicle against Geata in MS. Cot. Tib. A. 6, in Asser, and in the text of
Florence of Worcester (r, p. 71, ed. Thorpe, 1848), who, however, writes Geat
elsewhere : Beowan ham in the charter of 931 (Birch, Cart. Sax. n, 677,
Kemble, C. D. n, 353) as against Beaw in the Parker MS. : Hreftles [MS. Hradles,
Beowulf, 1485] as against Hreftlan [MS. Hrcedlan, 454] : Hors in the Historia
Brittonum, § 31, against the more usual Horsa. Some O.H.G. parallels will be
found in Z.f.d.A. xn, 260. There seems then no reason for doubting the
identity of Sceaf and Sceafa, or for regarding Sceafa as a shortened form of a
compound name with two elements (parallel to CuJ>a for Cu\>wine) as suggested
by Chadwick (282).
See also Koegel in Z.f.d.A. xxxvn, 270 : Mullenhoff in Z.f.d.A. vn, 410 :
Heusler in A.f.d.A. xxx, 31.
Longbeardum. The context in which the Lombards are here mentioned
seems to point to their ancient home between the Baltic and the Elbe.
The connection between the Longbeardan and the Heafyobeardan is doubtful.
See below, 1. 49.
33. Hun. The name of this chief is to be found in Beowulf, if we construe
11. 1143-4 with Bugge (P.B.B. xn, 33), Holder, Ten Brink and Gering " when
Hun thrust (or placed) in his bosom Lafing, best of swords." But the difficulties
of this interpretation were demonstrated by Imelmann in the D.L.Z. (xxx, 999 :
April, 1909). Chadwick (52) had drawn attention to the hero Hunleifus in the
Skioldunga gaga. E. Huchon, in a review of Chadwick in the Revue Ger-
manique (in, 626), pointed out that, on the evidence discovered by Chadwick,
Hunlafing in Beowulf must be the name of a man. This was confirmed by
Imelmann's discovery of Hunlapus (see Appendix, K). It may be therefore
taken as demonstrated that the name in Beowulf is Hunlafing, not Hun.
Possibly the name of a hero Hun is preserved in the Frisian Hunesga (Mullen
hoff in Nordalb. Stud, i, 160, Eieger in Z.f.d.A. xi, 188). The name Hunn
occurs in the list of sea-kings in the Thulor (C.P.B. n, 422). There does not
appear to be any connection between Hun of the Hsetwere and the king and
champion named Hun in Saxo. The name is not connected with the racial
name, being found as an element in personal names before the arrival of the
Huns in Europe. (Mullenhoff in Z.f.d.A. xi, 284.)
Hcetwerum. Leo first detected the Haetwere in this passage : Kemble had
read Hun-hcet Werum. The MS. may be read equally well either way. Grundtvig
(Dannevirke, 1817, n, 285) had identified the Hetware of Beowulf with the
classical Hatuarii : an identification which has been universally accepted. See
Grimm, DM. xxn; Zeuss, 100; Z.f.d.A. vi, 437 ; Mullenhoff, Beovulf, 17, 18,
94 ; Brerner in Pauls Grdr. (2) in, 894.
The Hatuarii were a Frankish people dwelling on the Rhine, near its mouth.
202 Widsith
Hringweald wees haten Herefarena cyning.
35 Offa weold Ongle, Alewih Denura :
Ofia ruled Angel, Alewih the Danes : he was boldest of all these men,
Zeuss would identify them with the Batavi of Tacitus, who were a branch of the
Chatti (Germ, xxix), but apparently the Hatuarii dwelt between the Rhine and
the Zuider Zee, whilst the Batavi dwelt south of them. They are first mentioned
as having submitted to Tiberius with other tribes of the lower Ehine (Vellejus,
ii, 105 ; subacti Canninefates, Attuarii, Bructeri). Subsequently they spread
somewhat higher up the Rhine, in the neighbourhood of Cleves, where they
were attacked by Hygelac in the sixth century, and overthrew him. (Beowulf,
2363, 2916 ; Gregory of Tours, Hist. Franc, m, 3. ) This district appears to be
designated by the O.N. Hunaland ; and Sigfrid, who in the Nibelungen Lied is
connected with it, is called inn hunski in the Edda (cf. Kaufmann in Z.f.d.Ph.
XL, 283-4). There may possibly be some connection between these terms and
the traditional king Hun of the Hastware.
Holen. Miillenhoff (Nordalb. Stud, i, 160) interpreted "elder-tree," com
paring for the naming of a hero from a tree the hero Assi in Lombard story [cf.
O.E. JEsc]: and for the sacredness of the elder comparing Grimm, DM. 374
(trans. Stallybrass, n, 651). Later he interprets it as a fictitious king's name,
"Holm-oak," emphasizing the idea of immemorial age (Z.f.d.A. xi, 284) and
compares O.H.G. hulis (modern hulst) used as a proper name. Neither of these
explanations is satisfactory, as the elder (Germ, holunder) is surely quite
distinct from the holly (O.E. holen) ; and it seems unlikely that O.E. holen
can mean " holm-oak."
Bugge (P. B.B. xii, 78) would identify Holen with the Holdinus rex Rutenorum
who is mentioned by Geoffrey of Monmouth as one of Arthur's barons (x. 6 :
x. 9). But this seems far-fetched.
Wrosnum. Mullenhoff (Nordalb. Stud, i, 161) compares the Rosogavi whose
inhabitants were rooted out by Charles the Great in 804. Et delude misit im-
perator scaras suas in Wimodia et in Hostingabi et in Rosogavi, ut illam gentern
foras patriam transduceret. Chronicon Moissiacense, in Pertz (fol.) SS. i, 307.
See also Bottger, Gaugrenzen Norddeutschlands, 1875-6, n, 144.
The people seem to have been coast tribes, dwelling between the Weser and
the Elbe, but nearer the latter. They would be close neighbours, if Mullenhoff's
conjectures are right, to the Woingas and the Ymbras, and not far removed
from the old home of the Lombards (Bardengau) or from Ymbria (Fehmarn).
The identification is therefore plausible, in view of the context.
Lappenberg's attempt (p. 176) to identify the Wrosnas with the Scandinavian
tribe who gave their name to Russia is, of course, chronologically and phoneti
cally impossible ; cf. Ettmiiller d) 19.
34. Hringweald, Herefarena cyning. Possibly, as Miillenhoff supposes
(Z.f.d.A. xi, 284), "epic fictions"; typical names for a treasure-giving king
and his conquering people.
Herefaran simply means the warriors or pirates. Mullenhoff (Nordalb.
Stud, i, 161) compares Lindesfaran : a band of sea-rovers who gave their name
to Lindisfarne. Moller (V.E. 90, 91) builds an elaborate theory on these similar
names, connecting the Lindesfaran or Herefaran, their king Hringweald, and
their Holy Island off the coast of Northumberland, with Heligoland, and the
neighbouring island of Sylt, where traditions of a king Ring remain.
35. Offa. See Introduction. Mullenhoff (Nordalb. Stud, i, 162) regards these
lines (35-49) as interpolated, although at an exceedingly early date. They do not
suit the context, and the praise of Offa, the foe of the Myrgingas, does not suit
a Myrging minstrel. Mullenhoff's other reason is that the historic Offa lived in
the sixth century, and therefore too late to have appeared as a hero of song
in the original Widsith. This last argument is certainly wrong (see Mullenhoff,
Beovulf, 85, 86). Later (Z.f.d.A. xi, 294) Mullenhoff does not try to distinguish
this passage, in date, from the bulk of the poem.
Ongle. Kemble suggested Onglum ; wrongly. Cf. 1. 8.
Text (It 34-41) 203
se waes bara manna modgast ealra;
nohwaebre he ofer Offan eorlscype fremede ;
ac Offa geslog aerest monna
cniht wesende cynerica msest ;
40 naenig efeneald him eorlscipe maran
on orette ane sweorde :
yet did he not in his deeds of valour surpass Offa. But Offa gained, first
of men, by arms the greatest of kingdoms whilst yet a boy ; no one of his
age [did] greater deeds of valour in battle with his single sword ; he drew
Alewih. Unknown. Thorpe (Cod. Ex. 517) followed by Miillenhoff (Nordalb.
Stud, i, 162) identified with the Olo, king of Denmark, frequently mentioned by
Saxo (Books vn, vin). But the names do not correspond phonetically, and
Miillenhoff subsequently gave up this explanation, suggesting another in his
Beovulf, 53. The name occurs in England in the Mercian pedigree (A.S.
Chronicle, Parker MS. sub anno 716) Alweo Sawing, & nephew of the great
Penda. Binz supposes the fact of a descendant of Offa having been named
Alweo to point to the survival of a tradition in which the two kings were
connected. Elsewhere the name occurs rarely in England and always north of
the Thames, on Anglian soil (P.B.B. xx, 169). The name is also fairly common
in continental Germany ; see instances in Forstemann, 54, 76 ; Alawig,
Alahwih. The O.N. form is Qlvfr (Kluge, Engl. Stud, xxi, 447).
37. nohwafre he ofer Offan eorlscype fremede. It has generally been assumed
that this means " Alewih could not exercise dominion over Offa " and that
it points to a struggle between the two kings. (Miillenhoff, Nordalb. Stud.
i, 162; Beovulf, 74; Moller, V.E. 30.) Eorlscype=(l) deed of valour, (2)
dominion, authority. The latter sense, however, is the rarer, and the phrase
Jremman or cefnan eorlscype seems always to mean " to perform deeds of
valour" ; e.g. in Beowulf, 2622, where it cannot mean to exercise dominion, as
Wiglaf is not a prince. Ofer may be [to rule] " over " as in rice ofer heofonstolas
"mighty over the thrones of heaven" (Genesis, 8) or "than," as loseph was
gleawra ofer hi ealle (Orosius, 34, 1). The rendering given above (which is the
one adopted by Koegel, Ltg. i, 1, 159) is therefore the more probable, though
both are grammatically possible. Bosworth-Toller gives both renderings, the
one under onorettan, the other under eorlscype. For ofer cf. F. Wuflen in
Anglia, xxxiv, esp. p. 448.
The allusion does not therefore necessarily point to any contest between
Danes and Angles.
38. geslog "gained by striking," as in the Brunanburg poem ealdorlangne
tir geslogon <et scecce : so in Beowulf (459) gesloh means "settled by blows."
Cf. Miillenhoff, Beovulf, 74. Not " he smote the greatest of kingdoms," viz.
Denmark, as taken by Moller (V.E. 31).
39. cniht wesende .. ,n<enig efeneald. But in both English and Danish story
Offa had reached the age of thirty before his combat. Mutus autem et verba
humana non proferens usque ad annum cetatis sues tricesimum ; De Offa primo in
Matthaei Paris Historia, ed. Wats, 1640. Qvi usqve ad tricesimum cetatis suee
annum fandi possibilitatem cohibuit, Sweyn in Langebek, i, 45 ; and cf. the
extract from the Annales Ryenses quoted on p. 87 (footnote).
Miillenhoff (Sagen, Mdrchen u. Lieder der Herzogthiimer Schleswig-Holstein
u. Lauenburg, p. 4) supposed that "thirty" was a blunder for "thirteen":
Chadwick (129) more plausibly suggests that such a phrase as }>ritig missera
may have been misunderstood as thirty years. But the extraordinary thing
is that the Danish and the English traditions should later agree in a common
error.
41. on orette. Thorpe (Cod. Ex. and Beowulf), Grein and Bosworth-Toller
suggest -that this should be read as one word onorettan, to accomplish, a word
204 Widsitk
42 merce gemserde wiS Myrgingura
bi Fifeldore : heoldon for?5 si]7)>an
Engle ond Swaefe, swa hit Offa geslog.
the boundary against the Myrgingas at Fifeldor. Engle and Swsefe held
it afterwards as Offa struck it out.
which, if we adopt the reading of Dietrich, Grein d) and Qrein-Wiilcker, is also
found in Exod. 312-3 : Judisc fefta an onorette uncuft gelad, " alone the folk of
Judah accomplished the strange path."
Although the editors have always read the MS. on orette, this might rather be
read as one word : hut little importance can be attached to the spacing of the
MS. Kenible adopted on orette, which his punctuation shows that he took as an
adverbial phrase "in battle" going with gemaerde. Ettmuller d) (2) suggested
cefnde on orette, "accomplished in battle": so Ebeling, Klipstein, Sedg.; Leo,
Grein (Sprachschatz), Bieger, Gr.-Wiilcker and Kluge also have on orette,
though Leo, Grein and Kieger certainly regarded orette as a verb (and con
nected with georfttan, orrettan ; which however means " to confound"). I follow
Kemble in the interpretation of on orette, and, with Bieger, make the pause
after sweorde.
oret is from *or-hat, a calling out, challenging : cf. O.H.G. urheiz.
ane sweorde. " Einzig mit dem schwerte, d. h. wohl durch zweikampf,"
Miillenhoff, Beovulf, 74. So Koegel, " allein mit dem Schwerte stellte er die
Grenze fest," Ltg. i, 1, 159. " With single sword he spread his borders,
against the Myrgings marked the bound," Gummere in M.L.N. iv, 420.
42. merce. mearce K (2), Ettmiiller d) (2) ; mcerce, Leo.
gemcerde. Fixed (not as Thorpe or Gummere take it " enlarged," " spread "
from mdra but) from gemsere boundary. Cf. lehtes singal tido gelimplicum
gimaerende, translating lucis diurna tempora, successibus diterminans — Kituale
Ecclesice Dunelmensis, 1840, p. 164.
43. Fifeldor. The combat, according to Saxo, took place on an island of
the river Eider (ed. Holder, iv, 115 ; xn, 402). The Eider is the Egidora,
JEgidora in the Prankish annalists of the 9th century (see Pertz (fol.) SS. i,
Einhardi Annales, 808, 811, 815, 828 ; Annales Fuldenses, 815, 857, 873) ;
jffigisdyr in the Old Norse (Jomsvikinga Saga, 8, 9). O.N. JEgir signifies both
the sea and the sea divinity, who was remembered in England till modern
times: he is the "monster Agar " of Lyly's Gallathea, and "eager" is still
used for the sea wave or bore of rivers. The O.E. is Egor (Eagor), which is
synonymous and interchangeable with Fifel : eagorstream (sea), Andreas, 441 ;
egorstream, Boethius, Metra, xx, 118; egor-here (deluge), Genesis, 1402, 1537,
alternate with Jifel-stream, Boethius, Metra, xxvi, 26 ; fifelwceg, Elene, 237.
Egordor and Fifeldor alike mean the mouth where the river opens on the dread
sea. This interchange of synonyms in a place name, although it is certainly
rare, can be paralleled from the same passages in the Jomsvikinga Saga where the
Eider is mentioned : the mouth of the Schlei being first called " mouth " and
then " door" (millom jEgisdura ok Slesmynna, cap. 8 ; Haraldr konungr fer til
jEgisdyra, en Hdkon jarl fer til Slesdora, cap. 9 ; Fornmanna Sogur, xi (1828),
pp. 28, 31. See Seelmann, Die Bewohner Danemarks vor dem Eindringen der
Danen, in the J.d.V.f.n.S. xn, 1886 ; and Bugge, H.D. 131-2).
Lastly, in the Chronicle of Dietmar, the Eider is called Wieglesdor, which is
best explained as a corruption of Fiflesdor (Thietmari Ghronicon, recog.
F. Kurze, Hannoverse, 1889, in, 6 (4), p. 51).
There would therefore, even apart from Saxo, be no difficulty in identifying
Fifeldor with the Eider : and, since Saxo definitely places Offa's combat on
that river, it seems carrying scepticism far to doubt the certainty of the identi
fication, as is done by Erdmann (p. 48), Holthausen and Siebs.
See also Grimm, D.M. 1835, 147, 197 ; Leo, 79. Bieger in Germania, in,
173 would connect Egidora, Fifeldor, with the Germanic idea of hell.
44. Swcefe. How do the Swsefe come in ? The contending parties are the
Text (U. 42-49) 205
45 HroJ?wulf ond HroSgar heoldon lengest
35 b sibbe IJaetsomne suhtorfaedran,
sij?ban hy forwraecon Wicinga cynn
ond Ingeldes ord forbigdan,
forheowan set Heorote HeaSo-Beardna J?rym.
For a very long time did Hrothwulf and Hrothgar keep the peace
together, uncle and nephew, after they had driven away the race of the
Vikings and humbled the array of Ingeld, hewed down at Heorot the
host of the Heathobards.
Angles and the Myrgingas. Therefore we must either (1) suppose that the
Myrgingas and Swrafe are here the same people, the Myrgingas counting as
one branch of the wide-spread Suevic stock. (Miillenhoff, Beovulf, 99-103 ;
Moller, V.E. 26.) Moller points to the name Schwabsted on the Trene, in
SchJeswig, as a proof that the Suevic family extended to the borders of the
Angles ; also the grouping of Engle and Sweefe together in 1. 60 supports
the idea that these races were neighbours. The fact that in 11. 23, 24 the
Swaefe are differentiated from the Myrgingas does not render this explanation
impossible ; we have seen that the term Swaefe from the time of Csssar and
Tacitus has a broader and a narrower signification. Or we may (2) take Sweefe
as a section of the Suevic tribe, isolated and settled around Schwabsted. Offa
perhaps rules both the Angles and this small tribe of Swabians : they are
bracketed together and opposed to the Myrgingas. This, certainly less plausible
theory, is supported by Koegel (Ltg. i, 160) and Seelmann (p. 57). (3) Rejecting
altogether the usual interpretation of Myrgingas as "people of Manrungani,"
Much (P.B.B, xvn, 193) would regard the name as synonymous with Saxon, and
would accordingly substitute Seaxan here for Sweefe. This view derives some
little support from Saxo Grammaticus, who makes the opponents of Offa
Saxons, but the conjecture is an improbable one.
Of the three theories (1) seems to present us with the fewest difficulties.
45. Hro]>wulf, Hroftgar, Ingeld. See Introduction, pp. 79-84.
lengest. See note to 1. 28, above.
47. Wicinga cynn. Siebs (p. 304), like Maurer earlier, regards this phrase
as indicating a late date. Wiking certainly in O.E. comes to mean a
Scandinavian pirate : but there is abundant evidence that, so far from being
a Scandinavian loan word, it existed in the West Germanic languages before
the Viking era. See Miillenhoff, Beovulf, 95, etc. ; Bugge, Studier over de
nordiske Glide- og Heltesagns oprindelse, i, 1881, pp. 5, 542 ; Much in P.B.B.
xvn, 201, etc. ; Sweet, History of English Sounds, p. 90. The original meaning
seems to have been " warrior," but the word came early to be connected with
the sea : cf. stewicingas, Exodus, 333.
49. Heafto-Beardna. It has been frequently argued that the true name of
the Langobardi was Bardi, and that hence the Heaftobeardan are identical with
the Lombards, the epithet being varied precisely as we get sometimes Dene,
sometimes Gar-Dene, or Hring-Dene. (So Ettmiiller (j) 20; Heyne's Beowulf;
Wiilcker, Kl. Dicht. 121 ; Bremer in Pauls Grdr. (2) in, 949.) In support
of this has been urged (1) the use of Bardi for Langobardi, in Paul the Deacon
and other writers in touch with Lombard tradition. But this use of Bardi for
Langobardi is almost limited to Latin verse. It is first found in the epitaph on
Droctulf , who was a contemporary of Alboin : an epitaph obviously written by a
Roman, not a Lombard (see SS. Rerum Langobardicarum in M.G.H., Hanno-
verse, 1878, p. 102). The form Bardi is used in several epitaphs written in
much later times by Paul the Deacon and others (see id. p. 22, on Arichis,
apparently by Paul ; p. 23 on Paul himself
Eximio dudum Bardorum stemmate gentis ;
the epitaph on Queen Ansa, attributed to Paul ; other epitaphs, 235, 238, 429).
It has not, I think, been noted that Langobardi is a word quod versu dicere non
206 Widsitk
50 Swa ic geondferde fela fremdra londa
In such wise I fared through many strange lands throughout this wide
est, and which had to be mutilated before it could be got into a hexameter. The
use of Bardi for Langobardi in verse proves, then, nothing at all. The only
early instances I can find of Bardus for Langobardus in prose are cohors Bardica
in the late 9th century Erchemperti Historia Langobardorum Beneventanorum
(SS. Eer. Lango. 262), and Bardan \>one beorh for the Apennines in Alfred's
Orosius (cf. Icel. Munbard, and Mons Bardonis in Otho of Frisingen).
Again, it has been pointed out (2) that men of Bardengau and Bardewik on
the Elbe are repeatedly called Bardi in the Chronica Sclavorum of Helmold :
but, granting that these men of Bardewik were a remnant of the old Lombard
stock, it seems perilous to argue, from the form of their name in the 12th
century, as to the form of the Lombard name, and consequent Lombard
affinities, in the sixth century. Moreover, it does not seem clear how the Bardi
remaining behind in Bardowyk could have been neighbours and foes of the
Danes : still less could the main body of the Lombards, who since A.D. 168 were
on the borders of the Roman Empire, by the Danube. If then we are to equate
the Heathobeardan and the Lombards, we must adopt the suggestion of Bugge
(H.D. 159-163) that a body of the Lombards was left behind on the Baltic coast.
But there is nothing to show that this was the case. Any evidence leading us
to identify the Heathobeardan and the Langobardi is therefore entirely wanting,
though of course the two nations may have been originally connected (cf. Binz
in P.B.B. xx, 174 ; Holler, V.E. 29). Miillenhoff (Beovulf, 29, 32, followed by
Much in P.B.B. xvn, 201; Heinzel, A.f.d.A. xvi, 271) suggested that the
Heathobeardan are rather to be identified with the Heruli (for whom see note
to 1. 87). Certainly the name wiring is most appropriate to the Heruli, who
were essentially sea-robbers : and they are the only people whom we know
to have had a blood feud with the Danes in the early ages. They were expelled
from their territory by the Danes, according to Jordanes. But, as the Heruli
are first heard of ravaging the Roman Empire in the second half of the third
century, it has been thought that their expulsion from their homes by the
Danes probably dates about A.D. 250 (Bremer in Pauls Grdr. (g) in, 834) ; whilst
the wars of Danes and Heathobeardan can be dated with some certainty about
A.D. 500. Yet since the Heruli who are met with in the Roman historians
about the end of the fifth and the early sixth century are still heathen, they had
probably only recently arrived from the North (Zeuss, 479), and it may have
been these sixth century Heruli, rather than the pirates of the third century,
whose influx into the Empire was due to their having been displaced by the
Danes. This would agree well with the chronology of Beowulf.
It seems probable that the Heathobeardan are personified in the hero
Hothbrodd (see Introduction, pp. 81-2) and since he is localized by the Helgi
lays on the south shore of the Baltic, and since this suits the data of Beowulf
and of Widsith well, we may perhaps regard this as the seat of the Heatho
beardan. More than this cannot be said, and evidence for identification with
any historic tribe, Langobardi or Heruli, is insufficient, though early kinship
with the Langobardi is probable enough.
See also Olrik, Heltedigtning, 21, 22 ; Sarrazin in Engl. Stud. XLII, 11.
d of Heaftobeardna added in MS. over line : cf. the spelling in Beowulf, 2037,
2067 ; Heaftabearna, Heaftobearna.
50. Swa. As the preceding passage has had nothing to do with the poet's
travels we must either (1) refer swa to what follows : I went through many strange
lands in such wise that... etc. (with Miillenhoff, Z.f.d.A. xi, 285) or (2) we must
suppose that the logic of the passage has been destroyed by omission or inter
polation (with Moller, V.E. 34 ; Langhans Veber den Ursprung der Nordfriesen,
Wien, 1877). Langhans regards the whole of 11. 10-49 as an interpolation and
would thus connect swa directly with the opening lines : "In such wise (i.e. accom
panying the lady Ealhhild)." In this Langhans may be right : but when he
argues that 18-49 are drawn from Beowulf, and from those parts of Beowulf
Text (II 50-58) 207
geond ginne grund; godes ond yfles
J>ser ic cunnade cnosle bidaeled,
freomaegum feor, folgade wide.
ForJ>on ic mseg singan ond secgan spell,
55 msenan fore mengo in meoduhealle,
hu me cynegode cystum dohten.
Ic waes mid Hunum ond mid HreS-Gotum
mid Sweom ond mid Geatum ond mid Su]?-Denum.
earth ; of good and evil there I made trial, from my race remote : afar
from my kindred, I served far and wide.
And so I may sing and tell my story ; declare before the company in
the mead-hall how men of great race were nobly liberal to me.
which have been most suspected of being late interpolations, his position
is untenable. See also Lawrence in Mod. Philol. rv, 335, 336, footnote.
53. folgade wide. I followed, served as a retainer. Eieger reads folgafte wide,
" far from my household," taking wide as equal to feor. But this is hardly
possible grammatically, and is certainly unnecessary. Equally unnecessary is
Kulipstein's/o^atyie wide ' in a wide retinue.'
57. Hreft-Gotum. See Appendix: Note H, The term Hrcedas applied to the
Goths.
58. Geatum. The Geatas of Beowulf. There must be some primseval con
nection between the names of the Geatas and of the Goths ; they stand in
ablaut relationship : Gaut- to Gut or Got. (See Bremer in Pauls Grdr. (2) in,
817, 818 : Erdmann Om folknamnen Gotar och Goter, Stockholm, 1891.) But
whilst the Goths were scattered over Europe, the Geatas (O.N. Gautar) have
remained throughout their whole history in the same locality : the country to
the south of the great lakes Wener and Wetter. Ptolemy is the first to locate
them definitely (n, 11, 16) xal Kar^xovaiv CU/TTJS [^Kavdias]. . .TO. 5e /j.effr)n^piva
FoOrai [read Fat/rat]. Procopius knows them in the same place (Bell. Gott.
n, 15). Already in Beowulf the Geatas are engaged in strife with the Swedes
dwelling on the other side of the great lakes : the strife resulted in the absorp
tion of the Geatas into the Swedish kingdom in prehistoric times (cf. Mullenhoff ,
Beovulf, 22 ; A.f.d.A. xvi, 269). In the earliest definite accounts of Sweden,
those given by the Christian missionaries, we hear only of one kingdom. But
from the middle of the llth to the middle of the 13th century warfare between
the two peoples was again constant, ending in the final absorption of the
Gautar. See Zeuss, 511-513 ; Bremer in Pauls Grdr. (2) in, 833 ; Fahlbeck (P.)
Den a. k. striden mellan Svear och Gotar, dess verkliga karaktar och orsaker, in
the Historisk Tidskrift, Stockholm, 1884, iv (pp. 105, etc.).
This identification of the Geatas and Gautar is etymologically exact, and
agrees well with the data of Beowulf: it has been accepted by most scholars
(e.g. EttmiiDer, Thorpe, Grein, Mullenhoff, Ten Brink, Sarrazin, Wiilcker,
Earle, Brandl, Grienberger, Schiick, Schiicking, Holthausen, etc.) and seems
as certain as almost any of the ethnological data of the Old English epic.
On the other hand Kemble wished to locate the Geatas in Schleswig, Grundtvig
in Gotland, and Haigh in England. The only other suggestion, however, j
which need be considered seriously is that of Leo, who wished to identify /
the Geatas with the Jutes. In this he was followed by Schaldemose, Bugge
(P.B.B. xn, 1 etc.), Fahlbeck, Gummere in M.L.N. iv, 421, and Gering (Beowulf).
The arguments in favour of this view will be found most fully in Fahlbeck's
Beovulfsqvadet sdsom kdlla for nordisk fornhistoria in the Antiqvarisk Tidskrift
for Sverige, vni, 2, 1. Fahlbeck points to a number of characteristics of the
Geatas as being more appropriate to Jutes than to Gautar : they are called
scemenn, brimwisan, 2930, 2953 ; their land ealond, 2333-5 : the Swedes reach
208 Widsith
59 Mid Wenlum ic waes ond mid Waernum ond mid
Wicingum.
them ofer sa, 2380 ; 2394, 2473. These arguments are inconclusive. The
strongest evidence in favour of the identification of Geatas and Jutes lies in the
use of Gotland, Gothos, for Jutland, Jutes by Alfred and Ethelwerd : and of
Eeffi-Gotaland for Jutland by the Icelanders : but this usage, though puzzling,
is not sufficient to establish Leo's hypothesis. See also pp. 237-241, 255.
Holt, adds ic wees after Sweom.
59. Wenlum. The same tribe as that referred to in Beowulf, 348, where
Wulfgar, Wendla leod is mentioned as dwelling at the court of the Danish king
Hrothgar. These Wen[d]le or Wen[d]las have been generally identified with
the Wendilenses of Saxo (Book xiv : called also Wandali, Book xi), the
inhabitants of Vendill, the modern Vendsyssel, in the north of Jutland.
Whether there is any connection between this tribe and the great East Germanic
race, the Vandals of history, is doubtful (see Bremer in Pauls Grdr. (2) in, 818).
Miillenhoff, however, would explain both the Wenlas of Widsith and the
Wendlas of Beowulf as referring to these Vandals of Eastern Germany
(Beovulf, 89) ; he thinks that Saxo comes too long after the time of Widsith
and Beowulf to justify us, without further evidence, in putting a chief of the
Jutland Wendilenses at the court of King Hrothgar : Wulfgar is a champion who
has migrated from the distant Vandals to the court of Hrothgar, rather than a
tributary prince of the neighbouring Vendill. There is nothing impossible in
Beowulf containing a reference to these Vandals, though they had migrated
long before to the South. The stories recorded by Paul the Deacon of combats
between Lombards and Vandals far in the North-East prove that the Vandals
were long remembered in tradition as dwellers in Northern Germany.
(Miillenhoff, Beovulf, 89-90. So Bugge, P.B.B. xn, 7; earlier Bugge had
followed Grundtvig in equating the Wendlas with the Wendilenses.)
Much in P.B.B. xvn, 210-11 rejects both identifications, and supposes the
Wendle to be neither the Vandali of Tacitus nor the Wendilenses of Saxo, but
rather a people akin to the Angles, and to those Cimbri who are noted by
Tacitus (Germania, xxxv) and Ptolemy (n, 11, 7) as inhabiting the north of
what is now Jutland. He supposes the name Wendle to have belonged in the
first place to these Cimbric ' Anglo-Frisian ' people, and only later to have been
taken over by the Danish tribe who had ousted or assimilated them. Wulfgar's
people then are half tributary to Hrothgar, but not yet Danish. Much urges
that the Wenle are here grouped with the Wasrnas, who were certainly a
non-Danish tribe akin to the Angles, and with the Vikings, a term which at
the time of Widsith would apply rather to Anglian than to Danish folk.
The trouble taken both by Mullenhoff and by Much to avoid identifying
the Wen[d]le of Beowulf and Widsith with the Danish inhabitants of Vendsyssel
is however quite unnecessary ; for we can gather from Procopius that about
A.D. 510 the northern part of the peninsula had been already settled by Danes
(Bell. Gott. n, 15). We know that later these Danes of North Jutland were
known as Vendilfolk; it seems clear then that the Wendla leod present at
Hrothgar's court must be their chief. The fact that they are mentioned here
in a non-Danish context cannot count for much : the poet wanted three W's to
alliterate (cf. Moller, V.E. 3, 4; Weiland, 15; Shuck, 43).
There was a famous Vendil north of Upsala (cf. K. Stjerna in A.f.n.F. xxi,
77, etc.) but there seems no reason to localize our Wen[d]las there.
Wicingum. Vikings. See 1. 47. Moller proposes WtiSingum, whom he would
identify with the Nuithones of Tacitus (Germ. XL), supposing the N to be
erroneous. These [N]uithones were a people akin to the Angles ; Moller would
locate them in the neighbourhood of Schleswig and supposes the name of the
river Widau (older Witha) and of the district Wiedingharde to have been taken
from this folk by the Danish and Frisian stock who supplanted them, and
who now occupy the district (V.E. 6-7).
Against this conjecture of Moller's see Much in P.B.B. xvn, 211.
Possibly the line is to be taken in connection with v. 62 and 68 : the last
Text (tt. 59-61) 209
60 Mid Gefjmm ic wees ond mid Wineduw ond mid
Gefflegum.
Mid Englum ic wees ond mid Swaefum ond mid
^Enenum.
name in each line not being the name of a tribe, but characterizing the preceding
names: "I was with Vandals and Wasrnas — Vikings: I was with Saxons and
Sycgan — swordsmen : I was with Franks and Frisians — brave men." But this
interpretation is very doubtful.
60. Gef\>um. The Gepidae ; near of kin to the Goths, and originally their
neighbours at the mouth of the Vistula. (See Jordanes, iv, xvn.) They
migrated south in the third century, and founded a kingdom in what is now the
S.E. of Hungary, which was overthrown after a protracted struggle by the
Lombards in the sixth century (567). See Hodgkin's Italvy, 132. They are .- U,
still known in Beowulf, though as a remote people J^WW); The mention of
them here together with the Wends, which is not necessitated by the allitera
tion, points to their position in the extreme east of Germania, bordering upon
the Slavonic peoples. See Introduction, p. 163, note 2.
Winedum. The Wends. These are the Veneti of whom Tacitus knew
(Germ. XLVI) though he was uncertain whether to reckon them as Germans
or not. The name seems to have been given by the Germans to their Slavonic
neighbours in the East. Jordanes (xxm) mentions the Veneti as having formed
part of the great empire of Ermanaric, and as pressing upon the German
nations in his day. Hermanaricus in Venthos arma commovit, qui, quamvis armis
despecti, sed numerositate pollentes, primum resistere conabantur... Nam hi...ab
una stirpe exorti, tria nunc nomina ediderunt, id est Venethi, Antes, Sclaveni,
qui quamvis nunc ita facientibus peccatis nostris ubique desceviunt, tamen tune
omnes Hermanarici imperiis servierunt (ed. Holder, 1882). The Wends pressed
forward into the lands between the Elbe and the Vistula vacated by the
Germans, where Wulfstan found them in the ninth century. (Alfred's Orosius,
ed. Sweet, 1883, p. 20.) See Grimm, G.D.S. 171, 190, 322; Miillenhoff D.A.,
1887, ii, 33 etc., 89 etc. For the correct form of the name see Collitz in the
Journal of English and Germanic Philology, vi, 282 (1906-7).
Gefflegum. Unknown. " Nach Skandinavien mochten noch zu setzen sein
die Geflegen, deren Name in Gefle, nordlich von Upsala, erhalten ist."
(Lappenberg, 176.) A rather desperate guess.
61. Englum. See Introduction.
Swcefum. See 1. 22 above.
jEnenum. L. Schmidt, Zur Geschichte der Langobarden, Leipzig, 1885
(p. 49), suggests a connection between the jEnenas and the Anthaib of Paul the
Deacon. But this is ignotum per ignotius, for no one knows to what place or
race Paul the Deacon was referring when he spoke of Anthaib ; neither did Paul
himself, who took the word out of the Origo gentis Langobardorum.
Grimm (G.D.S. 510) would connect the ^Enenas with a noble Bavarian race
the Aenniena or Anniona (see also R. A. 270) which is quoted in the Bavarian
laws as enjoying a double wer-gild (Leges Bajuuariorum, ed. J. N. Mederer,
Ingolstadt, 1793). Grimm sees in this line, which quotes together Angles,
Swabians, and (presumably) Bavarians, proof of a connection which he is
trying to establish between those peoples. See also Huschberg Geschichte des
hauses Scheiern-Wittelsbach pp. 51-61, especially p. 55, "Das Adelsgeschlecht
des Anniona war ansassig im Hochgebirge jenseits des Brenners " (between
Botzen and Trient).
Grimm's interpretation does not suit the context, as we are dealing with the
North Sea coast. Miillenhoff proposed to identify the ^Enenas with some tribe
of this coast [more particularly the Eowan] through the name of the Middle
High German hero Enenum, who comes from Westenlande, i.e. North
Friesland. This hero would be the last reminiscence of the forgotten ^Enenas
(Nordalb. Stud, i, 163).
c. 14
210 Widsith
Mid Seaxum ic wses ond [mid] Sycgura ond mid
Sweordwerum.
Mid Hronum ic wses ond mid Deanum ond mid
Hea]?o-Reamum.
Mid pyringum ic wses ond mid prowendum
65 ond mid Burgendum, |?ser ic beag ge]?ah :
And I was among the Burgundians; there I received an armlet.
62. Seaxum. See Introduction, pp. 67-69.
mid. Not in MS. or K, Leo. Supplied by Ettmuller (x) (2) silently, followed
by Thorpe and later editors ; though some (Thorpe, Grein, Eieger) are not
aware that mid is not hi the MS. Kluge retains the MS. reading.
Sycgum. See above, 1. 31.
Sweordwerum. See note to 1. 59. Lappenberg (178) suggested that we have
to do with the Suardones of Tacitus (Germ. XL), a people neighbouring upon,
and related to the Angles : though if the name is derived from sword as that of
the Saxons from sahs we should expect in Tacitus * Suerdones. See Grimm
G.D.S. 471; Much in P.B.B. xvn, 212 ; Weiland, 16 ; Seelmann Die Bewohner
Danemarks vor dem Eindringen der Danen, in J.d.V.f.n.S. xn (1886) 28;
Miillenhoff interpreted Suardones as Swordsmen (Nordalb. Stud, i, 119) but
changed his mind later (Z.f.d.A. xi, 286).
63. Hronum. Unknown. A suitable name for a sea-folk : cf. hron rod
(Miillenhoff Z.f.d.A. xi, 287). Apparently from the context a Scandinavian
people. Grimm (G.D.S. 751, following Ettmuller (x) 22) compares the Granii
who are mentioned by Jordanes (cap. iii) among the inhabitants of Scandinavia.
The word of course means " whale " and, says Moller, " cannot be a real tribal
name : people could only be called whales by friend or foe hi jest." ( V.E. 8.)
Cf. Kossinna in I.F. vn, 304.
Deanum. Obviously not the Danes, " doch weiss ich solche Deanas oder
altn. Daunir sonst nicht zu zeigen," Grimm G.D.S. 751. Miillenhoff proposed
to cancel the D, which he supposed to have crept in from the mid preceding,
and to read Eawum, the Aviones of Tacitus (cf. 1. 26 above), or else " if any
alteration is to be allowed " to read Heahum and to identify with the Chauci.
But in view of the scantiness of our sources Miillenhoff doubts whether the
reading may not be right, and preserve the name of some otherwise unrecorded
tribe. (Z.f.d.A. xi, 287.) As the context makes it probable that they are a
Scandinavian people, it is tempting to suppose that, by a mistake of an Old
English or a Greek copyist, they are the same people as are mentioned by
Ptolemy as AavKiuves (n, 11, 16), and located by him hi the south of Sweden,
but whom it has been impossible to trace in any other document.
Heabo-Reamum. Cf. Beowulf, 518, of Breca
\>a hine on morgentid
on Hea\>o-R(smas [MS. Bsemes] holm up <Etbcer,
They dwelt near the modern Christiania. Jordanes (cap. iii) refers to the
Eaumaricii or Raumaricice among the inhabitants of Scandinavia. The Norse
form of the word is Raumar or Hafta-Raumar, and the kingdom Raumarlki, near
Hafta land : the modern Bomerike and Hadeland. See Grimm, G.D.S. 751.
Brandl (992) suggests Baumsdal in Western Norway.
64. prowendum. The men of Drontheim (?), ON. prandir. See Lappenberg,
176; Grimm, G.D.S. 751. The sudden jump from Thuringians to men of
Drontbeim is strange : but we have had the Scandinavian Heatho-Beamas just
above (Miillenhoff, Z.f.d.A. xi, 288). Miillenhoff earlier (Nordalb. Stud, i, 165)
tried to get over the difficulty by interpreting pyringum as referring to the
Thuringians of the lower Bhine (see above 1. 30) and by referring prowendum to
an Old Frisian Threant. Brandl (967) identifies with the Treveri.
65. Burgendum. See note to 1. 19.
ge^ah. MS. gefreah.
Text (II 62-70) 211
me j?aer GuShere forgeaf glsedlicne maj?]min
songes to leane ; naes tycet sseue cyning !
Mid Froncum ic wses ond mid Frysum ond mid
Frumtinguwi.
Mid Rugum ic wses ond mid Glommura ond mid
Rurnwalum.
70 Swylce ic waes on Eatule mid JElfwine :
Guthhere gave me there a goodly jewel, the reward of my song. He was
no sluggish king ! I was with the Franks and the Frisians and the
Frumtings, the Rugians and the Glommas and the Romans. Likewise I
66. Gufthere. Thought of, perhaps, as at Worms. See Introd. pp. 58-63.
67. seene. Usually ' sluggish ' : here perhaps ' niggardly. ' This sense is
not elsewhere recorded in O.E., but the M.H.G. use of seine as "too short" (of
a coat) has been instanced. (Eichler in Anglia, Beiblatt xxn, 164.)
68. Froncum. See note to 1. 24.
Frysum. That these are the West Frisians, dwelling west of the Zuider
Zee, in the neighbourhood of the modern Amsterdam, is perhaps implied by
their being linked with the Franks, as in Beowulf 2912, 1207, 1210, where the
raid of Hygelac [who must have sailed up the Rhine] is repulsed by the joint
[West] Frisians, Franks and Attuarii. The Fresno, cynn of 1. 27, ruled over by
Finn Folcwalding, refers apparently to the main body of the Frisians, dwelling
east and north of the Zuider Zee. Heyne, in his index to Beowulf, notes that
the land of these North [or better East] Frisians, for they must not be confused
with the modern North Frisians of the islands, is called Frysland (1127), that of
the West Frisians Fresno, land (2916). In Widsith the e and y are reversed.
See Introduction, pp. 66-7.
Frumtingum. Nothing is known of them. Grimm (G.D.S. 752) would
interpret the word as a scornful nickname. Hullenhoff (Z.f.d.A. xi, 288) would
make it an " epic fiction " : to be taken here as an epithet of the Franks and
Frisians. Moller (V.E. 5) would make it equivalent to "pushing, brave folk."
69. Rugum, Glommum. See above, 1. 21, where the same people are
coupled together. But whereas in the earlier passage they were mentioned in
connection with a sea story, and evidently thought of as dwelling on the Baltic
coast, they are here mentioned in connection with the Komans, and are thought
of, apparently, as dwelling on the borders of the Empire, perhaps in what is now
Austria, which the Eugii occupied during the middle of the fifth century, till
their overthrow in 488 ; or perhaps in Italy itself, which they occupied with
the Ostrogoths. See Hodgkin's Italy, m, 169 ; Mullenhoff in Z.f.d.A. xi, 288-9.
Rumwalum. The Borne- Welsh, i.e. Romans. Moller (V.E., p. iii) proposed
to read Rumwarum, but this is an unnecessary alteration. The compound is a
natural one (cf. Wala rices below, and Gal-walum, the " Welsh " of Gaul, in the
A.S. Chronicle, sub anno 660). On the Franks Casket we have Romwalus and
Reumwalus for Bomulus and Remus : cf . Bugge, Studier over de Heltesagns
Oprindelse, p. 211.
70. Eatule. Leo (81) identified this as Italy. The identification is
certain : cf. the O.E. Bede, n, 4 (ed. Miller, p. 108) Ond he J>a ''Ses ilca papa
[Bonefatius] seono^S gesomnode Eotolwara biscopa. Miillenhoff (Z.f.d.A. xi, 289),
Moller (V.E. p. iii), Kluge and Holt, would read Eotule here. But ea for eo
occurs elsewhere in Widsith, cf. Earmanrices (1. 111).
The confusion of eo and ea is, of course, an Anglian, and more particularly
a Northumbrian peculiarity : cf. Sievers, Angelsdchsische Grammatik, 3te Aufl.
1898, § 150 (3). The West Saxon transcriber has removed this peculiarity in
the common nouns, but left Anglian peculiarities in proper names, which were
perhaps unfamiliar to him.
JElfwine. See Introduction, pp. 123-6.
14—2
212 Widsith
se heefde moncynnes mine gefrsege
leohteste bond lofes to wyrcenne,
heortan unhneaweste hringa gedales, ||
86 a beorhtra beaga, beam Eadwines.
75 Mid Sercingum ic waes ond mid Seringum.
Mid Creacuw ic waes ond mid Finnura ond mid
Casere,
se be winburga geweald ahte,
wiolena ond wilna ond Wala rices.
was in Italy, with 2Elfwine : he had of all men of whom I have heard the
readiest hand for deed of praise, a heart most liberal in the giving of
rings, of shining armlets — the son of Eadwine ! I was with the Saracens
and the Serings. With the Greeks I was and with the Finns, and with
Caesar, he who had rule over the towns of revelry, riches and joys and
the realm of Welsh-land.
72. wyrcenne. wyrceane, Holt.
74. Eadwine. See Introduction, pp. 123-4.
75-87. Brandl (Pauls Grdr. n, 1, 966) supposes this passage interpolated
by a half educated cleric who has drawn many of his names from Alfred's
Orosius. But, considering that very much the same ground is being covered,
the coincidences are certainly not striking. A heroic attempt has been made
by Mr A. Anscombe to identify many of the names in this passage with those of
tribes of Northern Gaul (hi Eriu, iv, 81-90, The Longobardic Origin of
St Sechnall). I am not able to agree with his conjectures.
For the character of this passage, see Introduction, pp. 7, 8.
75. Sercingum. Saracens (Lappenberg, 179; followed by Miillenhoff,
Z.f.d.A. xi, 289 ; who compare O.N. Serkir, and the O.H.G. gloss Sarzi,
Arabes). The ending ing is added on the analogy of English tribal names.
Other instances are Seringum (Seres), Lidwicingum (Letawicion, Letavici),
Exsyringum (Assyrians), Idumingum (Ydumaei).
Seringum. Seres (Lappenberg, 179). Syrians (Brandl).
76. Creacum. See above, 1. 20.
Finnum. See 11. 20, 79.
Casere. Here probably the Emperor of the West. Contrast 1. 20.
77. winburga. Holtzmann (Deutsche Mythol. herausg. Holder, 1874,
p. 185) would derive this old epic word from wyn, joy. The uniform spelling
win-, and the parallel medo-burg, leave us, however, no doubt as to the deri
vation from win, wine. Yet the spelling wynburga is followed by Leo, and
Ettmuller (a) (2).
78. wiolena ond wilna. The MS. has wiolane j wilna, which was kept by
Kemble d). J. Grimm (D.M. 1835, Appendix vi) recognized these words as
common nouns; in this he has been followed by Kemble (2), Ettmuller (j) (2),
Miillenhoff (Z.f.d.A. xi, 289), Bieger, Moller (V.E. iii), Wiilcker (Kl. Dicht.
glossary), Grein-Wiilcker, Kluge, Holt., Sedg. ; Leo kept the words as tribal
names Velena and Vylna translating " Der Walchen und Walchinnen tind des
Walchenreicb.es." The words are also taken as proper names by Grein, Thorpe
(Cod. Ex. and Beowulf), Brandl (Pauls Grdr. (2) 966-7) and Anscombe in
Anglia, xxxv, 527 : but no satisfactory explanation of them as such is forth
coming. The reading of the text, wiolena, is Rieger's.
Wala rices. See note on 1. 69, above.
Bugge sees an imitation of 11. 77, 78 in the Viribjgrg, Valbjorg of Gufrrunar-
kvfya, n, 33. See H.D. trans. Schofield, xxii; P.B.B. xxxv, 241.
Text (11 71-80) 213
Mid Scottum ic wses ond raid Peohtum ond mid
Scride-Finnum.
80 Mid Lidwicingum ic waes ond mid Leonum ond
mid Longbeardum,
79. Scottum, Peohtum. Miillenhoft (Z.f.d.A. xi, 290) thinks that the
mention of these names proves this passage to have been composed in England.
But a hard dwelling in the old Angel of Schleswig might have known of the
Picts and Scots by repute, as well as of the Finns and Hans.
Scride-Finnum. One section of the widely spread Finnish race, dwelling
more particularly in the north of Norway, and equivalent to Lapps rather than
Finns in the modern sense. The Finns mentioned before (1. 20) are, to judge
by the context in which they occur, the more southern section of the people,
dwelling east and north-east of the Baltic. The Scridefinnas take their dis
tinguishing title from sliding (Icel. skrifta) with snow-shoes as is explained by
Paul the Deacon. Scritobini...qui etiam ce&tatis tempore nivibus non carent, nee
aliud, ut pote feris ipsis ratione non dispares, quam crudis agrestium animantium
carnibus vescuntur; de quorum etiam hirtis pellibus sibi indumenta peraptant.
Hi a saliendo juxta linguam barbaram ethimologiam ducunt. Saltibus enim,
utentes arte quadam ligno incurvo ad arcus similitudinem, feras adsecuntur.
Paulus (i, 5, ed. Waitz, p. 50). Procopius gives an account of the ZtcpiOtyivot,
whom he numbers amongst the inhabitants of Thule [not Iceland, but
Scandinavia]. They are clothed in skins, and live by hunting: the mother
suspends her newly born babe, in a skin, from a tree : puts a fragment of
marrow in its mouth, and leaves it whilst she goes hunting with her husband.
But Procopius cannot vouch for the customs of Thule from personal experience :
'Ejiioi /j.tv ovv ^s Twurrfv Ui>ai rr)v vrjffov, TWV re elpr}fj.fruv ai>T6irTri yev4ff6ai, xalirep
y\iXOfJLtv<{>, T/)6ir(f> ovSevl ^w^x^7/- (Bell. Gott. n, 15.) Jordanes also mentions
the Scride-finnas among the inhabitants of the island of Scandia, but his scribes
have corrupted the name — Screrefennce, Crefennce, Rerefennte, possibly confusing
the Scride-finnas with another great branch of the family, the Ter-finnas.
The Geographer of Eavenna similarly gets confused over the name. In the
9th century the Scridefinnas are mentioned by Alfred in his Orosius (ed. Sweet,
1883, p. 16), be westannofftan him [the Swedes] sindon Scridefinnas and be
westan Norftmenn. Saxo speaks of the Scritfinni or Scricfinni (ed. Holder, 1886,
p. 8). See also Zeuss, p. 684, and some notes by Dahlmann in explanation of
Alfred's reference to the Finns (Konig JElfred's Germania), in his Forschungen,
i, 451-454 ; Bojunga in P.B.B. xvi, 547 ; Miillenhoff, D.A. n, 44.
80. Lidwicingum. E. Michel in P.B.B. xv, 377, etc. regards this name as
merely a variant of Wicingum (cf. Dene and Sa-Dene) ; and Bojunga, in his
criticism of Michel (P.B.B. xvi, 545), gives a silent consent to this view ; so
Grein, and Wiilcker iu Kl. Dicht. 129. But from the use of the name in the
A.S. Chronicle sub anno 885 we may be almost certain that the Armoricans, or
Letavici, are referred to. In that year is noted the accession of Charles the Fat
to the full domains of his great grandfather butan Lidwiccium (Parker MS.)
or Lidwiccum (Cotton Tib. A. 6). MSB. Tib. B. 1 and B. 4 have, however,
Lidwicingum. The form Lioftwicas is found in a later entry (Tib. B. 4 sub
anno 915) but Lidwiccas is the more usual form (cf. years 910, 918). In view
of the other passages in our poem in which the termination ing has been added
it seems clear that the reference is to the Lidwiccas. [The popular etymology of
the word is given in certain BISS. of the Historia Brittonum : the Britons who
settled Armorica slew the original inhabitants, and married their wives, having
first cut out their tongues lest they should corrupt the purity of their children's
speech : acceptisque eorum uxoribus et filiabus in coniugium, omnes earum
linguas amputaverunt, ne eorum successio maternam linguam disceret. Unde
et nos illos vocamus in nostra lingua Letewicion, id est Semitacentes, quoniam
confuse loquuntur. (Historia Brittonum, in Mommsen's Cronica Minora (M.G.H.),
in, 167, note.) The story is also told by Borrow in his Wild Wales and in
Skene's Celtic Scotland, 1886, i, 201 ; in, 96. See also Zeuss, 577-8.]
214 Widsith
8 1 mid HaeSnum ond mid Hserejmm ond mid Hun-
dingum.
Leonum. Dwellers in what is now Ostergotland, in the S.E. of Sweden.
They are probably the Aeuwvoi of Ptolemy : KO.I Kar^xovffiv aflrijs (i.e. SKCIJ'-
5/aj) TO. fdv dvriKCL "Kaideivol, rd. 8e /uea-jj/^Spti'ii. Tovrai Kal Aavxluves, ra de
fdffa Aev&voi (ed. Miiller, 1883, n, 11, 16). Ptolemy's account would thus put
the Haftnas in the west, the Geatas and the "Dauciones" in the south, and
the Leonas in the middle. The identification of the Leonas and the Aevwvoi
has, however, been disputed : some would amend the latter to Suewpoi, and
identify with the Suiones of Tacitus. (Bremer in Pauls Grdr. Q in, 830 ; see
also 818.) It is fairly clear that we must identify the Leonas with the
Lio-thida [thida = Goth. \>iuda, people] mentioned by Jordanes together with a
number of other pbscure tribes dwelling in Scandia ; and that they are identical
with the men of Ostergotland, whose capital is Linkb'ping, the old Liongkopungr,
and whose common parliament was known as the Lionga-]>ing. See Collin u.
Schlyter, Codex juris Ostrogotici (Ostgotalagen) in their Samling af Sweriges
gamla Lagar, 1831. Of. also Lappenberg, 176 ; Zeuss, 506 ; Miillenhoff in
Z.f.d.A. xi, 290. Miillenhoff, who regards the whole passage 75-87 as inter
polated, notes as remarkable the interpolator's acquaintance with Scandinavian
names, Leonum, Haftnum, HcBre&um.
Longbeardum. See 1. 32.
81. Hceftnum. Probably the XatSeivol of Ptolemy n, 11, 16, quoted in the
previous note (so Lappenberg, 175) ; the Heiftnir of Scandinavian history,
inhabitants of the Heffimork on the borders of Norway and Sweden. See
Bremer in Pauls Grdr. (2) in, 830; Much in P.B.B. xvn, 188; Zeuss, 159.
Leo, Ettmuller (j) (in his text, but not in his notes where he follows Lappen
berg), Moller (V.E. 8) take the word as a common noun " Heathen," but that
seems unlikely here. See also Bugge, P.B.B. xn, 10 ; Miillenhoff in Mommsen's
Jordanes, 165; Heyne-Schiicking, 120-316.
It is probably only an odd coincidence that what appear to be mutilated
forms of these names occur together in Beowulf (1982-4), where Hcereftes
dohtor bears the tankard Hce\j$]num to handa. See Sarrazin in Engl. Stud.
XLII, 17, 18.
ic was is added after Haftnum by Ettmuller, Thorpe (Beowulf), etc.,
Holt.
Heere\>um. MS. htsle\>um : the correction was made by Lappenberg (179)
followed by Ettmuller (j) (Haroftum in his notes, though he still reads htzlebum
in the text), Grein, Bieger, Moller, Wiilcker in Kl. Dicht, Holt, and
Sedg. But Ettmuller (2), Thorpe and Kluge retain the MS. reading. The
Herethas occur in the A.-S. Chronicle (Laud MS.) sub anno 787, On his dagum
comon (Brest HI scipu NortSmanna of Hereftalande. Dcet w<zron ]>a erstan scipu
Deniscra manna \>e Angelcynnes land gesohton. The Herethas slay the king's
reeve, who wishes to drive them to the town, because he knows not who they
are. These Herethas are the dwellers in the Norse HqriSaland on the Hardanger-
fiord. From their ravages Hiruaith comes to be the Irish name for Norway
(Zimmer in Z.f.d.A. xxxir, 205-6). There seem, however, to be another people
of the same name dwelling in Jutland : the inhabitants of HqrtS a Jotlandi, the
later Hardesyssel or Harsyssel. It is probably with this Jutland branch that
we must identify the XapoOSes of Ptolemy, who are clearly inhabitants of the
present Denmark. Ptolemy traces the tribes along the sea-coast till he comes
to the Cimbric Chersonese, where the XapoCSes dwell to the East, and the Cimbri
to the North : dvaroXiKurrepoi de XapovSes, irdvrwv 5' dptcriKwrepoi Kip/Spot, (n, 11,
7, ed. Miiller, 1883). The same people are mentioned with the Cimbri on the
Monument. Ancyranum : Cimbri et Chariides et Semnones et ejusdem tractus alii
Germanorum populi per legatos amicitiam meam et populi Eomani petierunt.
(See Suetonius, ed. Wolf, 2, 375.) For other references to the Haruthi see
Much in P.B.B. xvn, 203-5 : Zeuss, 152, 507, 519.
Hundingum. See 1. 23.
Text (II 81-86) 215
Mid Israhelum ic wees ond mid Exsyrmgum,
mid Ebreum ond mid Indeum ond mid Egyptum.
Mid Moid urn ic waes ond mid Persum ond mid
Myrgingum
85 ond Mofdingum ond ongend Myrgingum
ond mid Amothingum. Mid East-pyringum ic wses
82. Exsyringum. Miillenhoff suggests Assyrians (Z.f.d.A. xi, 291 ; Grein-
Wiilcker, i, 401). So Brandl.
Holt, supposes a name to have dropped out in this line.
83. Indeum. Grein reads Judeum by a slip of the pen. Holt, adds ic was
after Ebreum.
84. Moidum. Medes (Miillenhoff, Brandl as above). Holt, alters to Medum :
but cf. note by Sarrazin quoted in Vietor Festschrift, Marburg, 1910, p. 309.
Persum. This has always been taken as "Persians" and in the neighbour
hood of Moidum, " Medes," it is impossible to render it otherwise. Mr Anscombe
(Eriu, iv, 88) points out that Perse is used in the A.-S. Chronicle for Parisii,
sub anno 660. It is just conceivable that here Widsith is not so much inter
polated as corrupted ; that Persum and Idumingum may have stood in the
original poem, and have referred to Celtic and Baltic tribes : that, then, the
passage being imperfectly remembered or transcribed, someone proceeded to fill
up the gaps with Biblical names, which were suggested by the (apparently)
Biblical Perse and Idumingas.
Myrgingum. It is strange that the interpolator should place the poet's own
people, twice, amongst these exotic semi-biblical folk. Hence Miillenhoff (in
Grein-Wiilcker, i, 401) suggests that the people really meant here are the Myrce ;
the JElmyrcan and GuSmyrce of the Andreas and the Exodus.
85. Mofdingum. Mullenhoff gave up these people, and the Amothinge, in
despair, as "undinge" — perversions of some originally non-Germanic names
(Z.f.d.A. xi, 291). Later (in Grein-Wiilcker, i, 401) he identifies them with
Moabites and Ammonites. So also Brandl (p. 966). Grein suggested tentatively
mid Ofdingum (=? the Ubii). Lappenberg (179) suggested the Germanic
settlers of Moffat in Scotland !
Bieger reads Mid Mofdingum and mid Ongendmyrgingum.
ongend Myrgingum. Holt, suggests Mid Mofdingum ic wees ond mid
Mceringum : Sedg. adopts Maringum. Cf. Dear, 19 ; but this epic name of
Theodoric's retainers seems out of place in the same line with those of
Biblical peoples. Rieger and Gummere take Ongendmyrgingum as a compound
name. I should rather take ongend as a preposition ( — ongen, cf. Ongend\>eow
above) and translate "I was with the Myrgings...and I was against the
Myrgings " : an attempt of the interpolator to harmonize passages like 38-44,
where the poet is apparently opposed to the Myrgings, with 93-6.
85, 86 are metrically defective. They have been rearranged in a number of
different ways by the editors. The arrangement followed is that of Grein,
Grein-Wiilcker, Moller, and Kluge.
86. Amothingum. Grimm points out that the spelling (th for b) is sus
picious, and suggests Amolingas. The Amals were the ruling family among
the Ostrogoths (Vesegotha families Balthorum, Ostrogotlue prceclaris Amalis
serviebant, Jordanes, ed. Holder, cap. v). Hence the title Amulinga under
which Alfred refers to Theodoric, The name Amelunc remains in Middle High
German as one of his titles ; more generally, however, it is used with reference
to Theodoric's followers, and this is the meaning which Giimm would give to
the word here (G.D.S. p. 598). Lappenberg (179) identified with the Othingi
of Jordanes (in, 21) followed by Ettmuller d) 24 and Thorpe (Cod. Ex. 520).
Nothing is to be said in favour of this. Ettmuller rearranged the line mid
Amothingum ic vces and mid East\>yringum ; similarly Holt., but reading
Amoringum, "Amorites." See also above, 1. 85.
216 Widsith
87 ond mid Eolum ond mid Istum ond Idumingum.
Ond ic wses mid Eormanrice ealle )>rage,
And I was with Ermanaric all the time : there the king of the Goths
East-\>yringum. The Thuringians, called East-Thuringians, either (1) to
distinguish them from the people of the same name dwelling in Zeeland and
North-Brabant (see above, 1. 30), or (2) as indicating that portion of the
Thuringians which in the sixth century was still dwelling east of the Elbe ;
as suggested by Seelmann (Nordthttringen in J.d.V.f.n.S. xn, 1886, p. 1).
MiillenhofFs later interpretation of the name (Grein-Wiilcker, i, 401) as a
corruption of "Assyrian" is unlikely.
87. Eolum. The name cannot be identified, and is probably corrupt.
Grimm, apparently by an oversight, offered two different conjectures in the
G.D.S. : (1) he suggested (p. 598) the emendation Eorlum, interpreting the
(H)eruli, *E/jouXot. Linguistically this would be quite satisfactory. Corre
sponding to Eruli we should expect O.N. *Jarlar ; O.E. *Eorlas : the name
is probably connected with the common noun eorl "lord": the "u" in the
classical form Heruli is not original (Moller in A.f.d.A. xxn, 152, 160). This
etymology is to be preferred to the derivation from heru " sword," supported by
Grimm (G.D.S. 470) and Erdmann (77). (2) ,In G.D.S. 736, Grimm suggested
Eotum ; these would be identical with the Ytum of 1. 26 (q.v.). Miillenhoff (in
Grein-Wiilcker, i, 401) suggests that the name is a corruption of Elam or Elath,
Brandl of Eowland, Holt, of Aeoles, Aeolians.
Grimm's first identification, with the Heruli, is the more probable. These
seem to have been a Scandinavian people, southern neighbours of the Geatas.
They also occupied at a very early date portions probably of the south coast of
Scandinavia and of the adjacent islands, from which they were driven by the
Danes. (Dani...Herulos propriis sedibus expulerunt, Jordanes, in.) The Heruli
seem to have wandered over Europe, and to have been engaged as mercenaries
in many places. Jordanes repeatedly refers to their swiftness and value as light-
armed troops. Procopius (Bell. Gott. n, 14, 15) gives a less favourable account
of them. Of old they had been accustomed to put to death their sick and aged,
and to compel widows to commit suicide ; and in spite of their becoming allies
of Borne and nominally Christians, they remained dirurroi. ical ir\eoi>f£iq. tx6-
fj.fi>oi...trovr]p&Ta.Toi avOpdnruv aTrdvrwv. After their overthrow by the Lombards,
a party of them returned to Thule (Scandinavia). They disappear from history
in the sixth century : and the absence of any mention of them in Beowulf, the
scene of whose actions is laid near what had been their territory, is noteworthy
[but cf. Miillenhoff, Nordalb. Stud, i, 124-5]. Miillenhoff (Beovulf, 30-32) would
identify them with the Heatho-beardan.
Note that Theodoric addresses together the kings of the Thoringi and Heruli
(Varite, in, 3). See Seelmann, Das Norddeutsche Herulerreich in the J.d.V.f.n.S.
1886, pp. 53-7.
Istum. See Appendix, Note F.
Idumingum. See Appendix, Note G.
Ettmiiller, Grein and Bieger (who transposes Eolum and Eastty/ringum and
assumes a gap after Istum) insert mid before Idumingum. Holt, reads : mid
Eolum ic wees ond mid Istum ond mid Idumingum. I follow the MS. closely
throughout all this passage. The evidence which it shows of an unskilled
hand may throw light upon the history of the poem, and should therefore be
retained.
88. Thorpe suggested that ealle ]rrage should be taken in close connection
with gode dohte, "all which time to me the Gothic king was bounteously kind."
This gives sense : Eormanric was treacherous and hasty, and might well not
have been consistently bounteous all the time Widsith was with him. The
recognized punctuation would give "I was with E. all the time," the meaning
of which is not clear ; and if this punctuation is kept we must agree with
Lawrence (Mod. Philol. p. 359) that the line does not follow well on the pre
ceding line, nor yet make such a faultless connection with 1. 74 as Miillenhoff
Text (II 87-101) 217
}*er me Gotena cyning gode dohte;
90 se me beag forgeaf, burgwarena fruma,
on J?am siex hund waes smaetes goldes
gescyred sceatta scillingrime ;
J?one ic Eadgilse on aeht sealde,
minum hleodryhtne, )>a ic to ham bicwom
95 leofum to leane, J?aes J?e he me lond forgeaf,
mines faeder ej?el, frea Myrginga.
ond me Ipa, Ealhhild o)>erne forgeaf,
dryhtcwen duguj?e, dohtor Eadwines.
Hyre lof lengde geond londa fela,
too ]>onne ic be songe secgan sceolde,
hwser ic under swegl[e] selast wisse
was bounteous unto rne. Lord of cities and their folk, he gave me an
armlet, in which there was reckoned of refined gold, six hundred pieces
counted in shillings. This I gave into the possession of my lord and
protector Eadgils, when I came home: a gift unto my beloved prince,
because he, lord of the Myrgings, gave me my land, the home of my
father. A second ring then Ealhild gave unto me, noble queen of
chivalry, daughter of Eadwiue. Through many lands her praise extended,
when I must tell in song, where under the heavens I best knew a queen
maintained (Z.f.d.A. xi, 291) : something has therefore in that case presumably
been lost, as well as something interpolated. Thorpe's rendering appears to me
forced, and I doubt if we can take \ar with ealle \rage, as suggested by Eichler
in Anglia, Beiblatt xxn, 164 : " alle die Zeit iiber, in der mien der Gotenkonig
beschenkte." See Introduction, p. 21 etc.
91-2. The sceatt is sometimes a fixed sum : a subdivision of the shilling :
but often its meaning is quite vague, "a piece of money." It is best so taken
here, the value of the sceatt being then defined by scillingrime. See Appendix,
Note N, On the beag given by Eormanric.
gescyred ; reckoned, numbered, as in the Pharao, of Pharaoh's chariots
\&t \ar screoda ware gescyred rime
siex hundreda searohcebbendra .
So here "in which there was reckoned of refined gold, six hundred pieces
counted in shillings," i.e. the ring was worth six hundred shillings (so Koegel,
Ltg. i, 1, 139). It is not necessary to press the passage further, or to suppose,
with Clark Hall (Beowulf, 1901, p. 177) and Stopford Brooke (r, 3), that the
ring was scored into six hundred sections, each of the weight and value of a
shilling.
beaghord for beag, the suggestion of Grein, and on scillingrime, the suggestion
of Ettmiiller (1), are unnecessary alterations of the text.
95-6. The land had not necessarily, as Koegel supposes (Ltg. i, 1, 139),
been forfeited or lost : but on the death of the father it would have needed to
be granted anew to the son (cf. Beowulf, 2607, etc.). This is all that is neces
sarily implied in Widsith's statement that his lord gave him his father's
inheritance.
96. Ettmiiller d) places a full stop after Myrginga, rightly : Ettmiiller (2)
a semicolon. See Introduction, p. 27.
101. swegle. MS. swegl, followed by K., Thorpe (Cod. Ex.) ; swegle, Leo,
etc., rightly.
218 Widsith
goldhrodene cwen|| giefe bryttian.
Bonne wit Scilling sciran reorde
for uncrum sigedryhtne song ahofan,
105 hlude bi hearpan hleo)>or swinsade,
}>onne monige men modum wlonce
wordum sprecan, )?a ]?e wel cu)>an,
Ipcet hi nsefre song sellan ne hyrdon.
Donan ic ealne geondhwearf e]?el Gotena,
no sohte ic £ [ge]si)>a ]?a selestan :
]?aet wses innweorud Earmanrices.
HeScan sohte ic ond Beadecan ond Herelingas
adorned with gold giving forth treasure. When Scilling and I with clear
voice raised the song before our noble lord (loud to the harp the words
made melody) then many men cunning and great of mind said they had
never heard a better song.
Thence I wandered through all the land of the Goths : I ever sought
the best of comrades, that was the household of Ermanaric. Hethca
I sought and Beadeca, and the Harlungs, Emerca and Fridla, and East-
102. giefe. Thorpe (Beowulf) suggests giefa : unnecessarily.
103. Donne. MS. don, followed by Kemble, Schaldemose. Later edd., Leo,
Ettmiiller d) (2), Thorpe, Bieger have altered d to ft silently.
Scilling. The name survived in England till historic times. A Scilling
presbyter signs charters in Wessex during the third quarter of the eighth century
{Birch, Cart. Sax. i, 186, 200); as does a contemporary Scilling, prefectus
(Birch, i, 224, 225), also in Wessex : a Scilling also witnesses a Mercian charter
(Birch, i, 181) of approximately the same date.
106-7. Moller (V.E. 10) condemns these verses as " wooden," and supposes
them expanded by an interpolator from the simple
]>onne wordum sprecan wlonce monige.
Granting the woodenness, we have no right to assume that authentic O.E. poets
never wrote anything wooden.
107. sprecan. Leo, Ettmiiller d), unnecessarily normalize to the W.S.
sprcecon ; e for OK is simply a survival of tbe original Anglian dialect of the
poem.
108. sellan. scelran, Leo, Ettmiiller (j) : selran, Ettmiiller (2) : sellan song,
Holt, for metrical reasons.
109. Donan, i.e. from the hall of Ermanaric. See Introduction, p. 53, etc.
Moller (V.E. 2-3) supposes these lines in the first place to have followed im
mediately after 11. 88 and 89, and Scwum to refer to "Seer. Widsith first visits
the court of Ermanaric and there receives a ring : thence he voyages through
the whole Gothic land.
110. gesfya. The MS., followed by Kemble and Thorpe, has sfya, " of courses
I ever sought the best," so Wulcker in Kl. Dicht. ; Leo siftlSan ; Ettmiiller (i)
gesVSa, followed by Grein, Eieger, Grein- Wulcker, Kluge, Holt., Sedg.
110 is clearly to be taken in close connection with 1. 109 : the semicolon
inserted at the end of that 1. by Grein, and followed by Wiilcker, is better
replaced by a comma, as suggested by Mullenhoff, Z.f.d.A. xi, 291.
111. Earmanrices. Kemble, Leo, Ettmuller, Thorpe, Grein and Holt, would
alter Ea to Eo. But cf. Eatule for Eotule in 1. 70.
112. Heftca. Unknown (Binz in P.B.B. xx, 153). Mullenhoff (in Beovulf,
1, 64) speaks as if Heftca came into tbe East Saxon genealogy. This however
is not the case: only Biedca, Bedca appears there. Sarrazin (Anglia, ix, 202)
Text (U. 102-116) 219
Emercan sohte ic ond Fridlan ond East-Gotan,
frodne ond godne feeder Unwenes.
115 Seccan sohte ic ond Beccan, Seafolan ond peodric,
Heaj?oric ond Sifecan Hli]>e ond Incgenbeow.
Gota, sage and good, the father of Unwen. Secca I sought and Becca,
Seafola and Theodric, Heathoric and Sifeca, Hlithe and Incgentheow.
suggests that Beadeca and HeSca may be identical with the Boftvarr and
Hgttr of the saga of Eolf Kraki. This seems an exceedingly improbable
conjecture.
Beadeca was a well-known hero in England : his name is associated with
lea, well, and hill (see Binz in P.B.B. xx, 152-3), and, in the form Bedca,
comes into the East Saxon genealogy, as given by Florence of Worcester (ed.
Thorpe for English Historical Society, 1848, i, 250), and as Biedca into Henry
of Huntingdon (Historic Anglomm, 11, § 19, ed. Arnold, 1879, Bolls Series,
p. 49). See also Grimm, D.M., Anhang, Patuhho.
The vowel is short ; not Beadccan, as Grein, Ettmiiller (j).
Herelingas, Emercan ond Fridlan. See Introduction, pp. 28-36.
113. sohte ic. Mullenhoff (Z.f.d.A. xi, 291-293) regards the repetition of
sohte ic as due to interpolation. It certainly seems wrong here, for Emerca and
Fridla are the Harlungs, and the two words were therefore rightly deleted by
Ettmiiller d) (2). But Miillenhoff would also strike it out in 11. 112, 115, 117,
119, 123, on the ground that it overfills the line. This however would mean
that sohte ic would have to be understood in 1. 123 from 1. 110, in spite of a
parenthesis (119-122) of several lines intervening. To avoid this Mullenhoff
would take the names in 123, 124 not as accusatives after sohte ic [understood]
but as nominatives in apposition to Hrceda here : altering Wudgan and Haman
into Wudga and Hama.
East-Gotan. Bojunga and Wiilcker (Kl. Dicht. glossary) have taken as
plural, "Eastgoths" : but there can be no doubt that this is a reference to the
hero Ostrogotha, for whom see Introduction, pp. 13-15.
114. Unwenes. Jordanes (xiv, 79) has " Ostrogotha autem genuit Hunuil:
Hunuil item genuit Athal," upon which Mullenhoff comments Hunuil nominis
certe ultimam partem corruptam esse nemo non videt. He suggests in view of
this passage in Widsith that Unuin should be read : Unwin stands for Unwen as
Thiudimir interchanges with Thiuderner. The name means a son born beyond
hope, cf. Z.f.d.A. xn, 253. Unwen must have been a famous hero to have
been remembered in England till after the Norman Conquest (see Appendix K).
Yet he does not succeed his father Ostrogotha in Gothic story : he was
therefore presumably cut off in his youth. Was it in this connection that
Ostrogotha, like Thurisind or Olaf the Peacock, showed his characteristic
patience ?
115. Seccan. See Mullenhoff in Z.f.d.A. xi, 276.
Beccan. See above, 1. 19.
Seafolan ond peodric. See Introduction, pp. 40-44.
Seafola probably signifies "the cunning one" from a root preserved in
O.E. sefa, " understanding," or Latin sapere. (Mullenhoff in Z.f.d.A. xxx, 240 ;
Much in P.B.B. xvn, 199.) The names of the traitors Seafola and Sifeca are
thus perhaps connected.
For further notes on Sabene see Grimm's Heldensage, 235 ; Bugge, H.D.
passim.
116. Hea.\>oric. Miillenhoff (Z.f.d.A. vr, 458) suggested that this hero is
the O.E. equivalent of the Frederic who in the continental German versions is
murdered by his father Ermanaric. But this is a quite unnecessary assumption,
as a Freo\>eric is mentioned below.
Binz (P.B.B. xx, 208) and Schutte (A.f.n.F. xxi, 37) would identify HeaXoric
with Heiftrekr, the ferocious king who is the half-hero of the Hervarar Saga.
220 Widsith
117 Eadwine sohte ic ond Elsan, ^Egelmund ond Hungar
Eadwine I sought and Elsa, JSgelmund and Hungar and the proud
But the two names do not correspond, and though the Old English one may be
corrupted, we can hardly regard this as more than a probable guess. Schutte
would further identify Heaftoric-HeiSrekr with the historic Ardaricus (Jordanes,
ed. Mommsen, L). Lappenberg in 1838 had suggested "Heathoric der Haidreck
von Hunaland im Oddruasliede [Oddriinargrdtr] in der Edda sein kann" (p. 181).
See also Much in Z.f.d.A. XLVI, 315.
Sifecan. Clearly the traitor Sibich, for whom see Grimm, Heldensage, 18, 19.
Binz's conjecture seems improbable. See Introduction, p. 34, footnote.
Hli]>e. Generally identified with Leth, the third king of the Lombards.
See Introduction, p. 122 (first by Lappenberg : supported by Ettmuller d) 25 ;
Miillenhoff, Z.f.d.A. xi, 278; Moller, V.E. 11).
It was pointed out however by Svend Grundtvig in 1856 that Hlfye and
Incgen]>eow probably correspond to the Hlgftr and Angantyr of the Hervarar
Saga. (Danmarks gamle Folkeviser, 4°. Kjobenhavn, 1856, n, 637.) Hloth and
Angantyr are half-brothers and foes ; but as they are both concerned in the
great fight between Goths and Huns which is obviously connected with the
struggle referred to in 11. 120-122, they naturally have a place here. See
Introduction, pp. 46-48.
Incgen]>eow. If this is Angantyr, we should rather expect Onijerfyeow. For
the present form however compare Ingceburne, Incgeneeshdm in Kemble, Codex
Dipl. 813 (iv, 157), 593 (m, 127). See Ongend^eow, above, 1. 31.
117. Eadwine (Audoin) and JSgelmund, below, are very possibly Lombard
kings. Moller (V.E. 11) regards the Lombard passage as an interpolation.
This seems more likely than that it points to a contamination of Lombard and
Gothic story, centering perhaps in the person of Ealhhild-Swanhild. See
Introduction, p. 122.
Elsa. Coming between Eadwine and JEgelmund, Elsa ought to be the name
of a Lombard hero, and it is frequently asserted that it is (e.g. by Binz in P.B.B.
xx, 206, Lawrence in Mod. Philol. iv, 344). For this there is no evidence except
that the name Alisso occurs in a charter drawn up at Beneventum under
Lombard rule in (?) 752. (See Meyer, Sprache der Langobarden, 1877, p. 195 :
Troya, Codice diplomatico longobardo, 1852-54, iv, 440, No. 668.) But there is
no proof that this Alisso was a Lombard : he is a Condoma, and, whatever that
word may mean (see Ducange, Condamina, where the charter is quoted), it is
clear from this charter alone that many persons of this status were not
Lombards, or even Germans. Even were Alisso a Lombard, it does not follow
that he was named after an early Lombard king or hero. The statement that
Elsa is a Lombard hero is therefore unsupported conjecture.
On the other hand the name has come down to us as part of the Gothic
tradition of Dietrich von Bern, in whose cycle there occur two heroes of the
name Else, father and son (Biterolfu. Dietleib, passim; Dietrichs Flucht, 8313).
There is also Elsan, der alte, der guote to whose care the young princes are
entrusted in the Eabenschlacht, and who has to atone with his life for his
neglect of his charge. Also we have an Elsung in the Thidreks Saga. Evidence
would therefore point to Elsa being a Gothic hero. The name Elesa however
occurs in the West Saxon pedigree as that of the father of Cerdic (for a dis
cussion of the name see Kemble, Stammtafel der Westsachsen, 1836, p. 27) :
whilst Alusa occurs in the allied Northumbrian pedigree, as given in the early
ninth century genealogies (Sweet, Oldest English Texts, 1885, p. 170, and in
Florence of Worcester, ed. Thorpe, 253). See Mullenhoff, Beovulf, 63 ; Grimm,
Heldensage, 138, 192. For Alisso see Zacher, Das Gothische Alphabet Vulfilas,
Leipzig, 1855, p. 106 ; for a possible derivation and cognates of Aliso, Elisa,
Ihung see Miillenhoff, Ein Altsdchsische Gott Welo, in Nordalb. Stud, i, 36.
jEgelmund. See Introduction, pp. 121-3.
Hungar. Mullenhoff (Z.f.d.A. xi, 284) would identify him with Onegesius,
the viceroy of Attila, who is constantly mentioned in Priscus' account of his
Text (II 117-121) 221
ond ]?a wloncan gedryht Wty-Myrginga.
Wulfhere sohte ic ond Wyrmhere : ful oft J?ser wig
ne alaeg,
120 Jjonne Hrseda here heardurn sweordum
ymb Wistlawudu wergan sceoldon
company of the With-Myrgings. Wulfhere I sought and Wyrmhere:
there full oft war was not slack, what time the Goths with sharp swords
must defend their ancient seat from the people of Mtla. by the Vistula-
wood.
embassy (see Miiller, Fragmenta Hist. Greec. iv). It has been denied that
Onegesius was a Goth. Thierry (Histoire d'Attila, nouvelle e"dit. 1864, p. 93)
concludes that he was a Greek, whilst Hodgkin (Italy, 11, 74 note) argues that
he was a Hun, and orientalizes his name into Onegesh. Oiiegesius acts as
intermediary and apparently as interpreter between the ambassadors and Attila.
But in the Acta Sanctorum (Parisiis, 1868, July 29, p. 81) Attila's interpreter
Hunigasius [sic, not Hunigaisus] is mentioned. The two are almost certainly
identical, as Hodgkin admits (n, 123). But Hunigaisus is a purely Gothic
name, and corresponds exactly to Hungar. As a victorious Gothic leader who
had overcome the Acatiri (Priscus, pp. 82-3, 86), sitting on Attila's right in
hall (91), fj.era rbv 'A.rr-f)\o.v irapb ZtdjOais Iff-^ixav fjAya. (85), Hunigais would be
very likely to have passed into Gothic saga. He may well be the Hungar here
referred to.
118. Wi\>- Myrginga. wi% Myrginga, Kemble, Leo, "the company against
the Myrgings " ; winas Myrginga, Ettmuller (j) ; wine Myrginga, Ettmiiller (2) ;
Wi\>-Myrginga, Thorpe (Cod. Ex. and Beowulf), Kieger, Grein, Wiilcker in Kl.
Dicht., Grein-Wiilcker, Kluge, Holt., Sedg. Unless, as Heinzel (Hervararsaga
in W.S.B. xiv, 101) suggests, these are the Lombards, we must abandon the
name as inexplicable. The Lombards might possibly have been also called
Myrgingas, for in all probability they occupied Maurungani and its neighbour
hood for three centuries (from the middle of the second to the end of the sixth),
"hovering about the skirts of the Carpathians, perhaps sometimes pressed
northwards into the upper valleys of the Oder and the Vistula" (Hodgkin,
Italy, v, p. 102). Mullenhoff (Nordalb. Stud, i, 149) interprets Wfy-Myrginga,
"Friends of the Myrgings," i.e. Lombards [the same interpretation had been
given by Ettmuller u),p. 25]. Mullenhoff comments, "There is a very close
connection postulated by the poem between the Lombards and the Myrgings."
So Gummere, " neighbor- Myrgings " (cf. Moller in A.f.d.A. xxu, 152).
To suppose any connection with the Icelandic Myrk-viftr (as Schutte,
A.f.n.F. xxi, 34, 37, and Sarrazin, Engl. Stud, xxin, 236-7) is surely impossible.
We should expect, if anything, the O.E. equivalent, not a dislocated Icelandic
word. But see also Schutte, Oldsagn, 63.
Wulfhere. Unknown, although both Ermanaric and Theodoric have, in
story, retainers whose names begin with Wolf. Cf. Biterolf and Alpharts Tod,
passim, and Wolfram dietrich in the Low German King Ermanaric's Death, etc.',
and see Grimm, Heldensage, 238 ; Symons in Z.f.d.Ph. xxxvm, 161 ; Matthaei
in Z.f.d.A. XLVI, 55 ; and note on Wulfingum, above (1. 29).
119. Wyrmhere. "Le nom Vyrmhere re'pond tout-a-fait & Ormarr de la
Hervararsaga, tout comme Gufthere, en ancien allem. Gundahari, repond &
Gunnarr" — C. C. Bafn in Antiquites russes, Copenhague, 1850, i, 112.
This identification seems fairly certain, and supports, and in its turn is
supported by, the interpretation of Hlfye ond Incgerifreow as Hlf&r and An-
gantyr, as suggested by Svend Grundtvig and Heinzel.
120. Hrceda. Ettmuller, Holt, change to Hrefta, unnecessarily, cf. Mullen
hoff, Z.f.d.A. xn, 259, and see Appendix : Note H, The term Hraedas applied to
the Goths.
121. Wistlawudu. The Goths left the Vistula towards the end of the
222 Widsith
ealdne ebelstol ^Etlan leodum.
Kaedhere sohte ic ond Rondhere, Rumstan ond
Gislhere,
Wtyergield ond Freoberic, Wudgan ond Haman :
125 ne waeran ]>cet gesiba ba ssemestan,
Not the worst of comrades were they, though I am to mention them
second century A.D. (Hodgkin's Italy, i, 40). These lines therefore preserve
a very early tradition, but we can draw no exact chronological argument from
the allusion except that we are here dealing with saga and not with history.
See Introduction, p. 163. The wood is probably to be identified with the
Mirkwood which, in later Icelandic story, separates Goths and Huns.
There is a mark in the Exeter Book over the a, like an imperfectly formed
mark of length : it is quite unlike the curved heavy horizontal stroke which
signifies n or m. Mr Anscombe's suggestion (Anglia, xxxiv, 526) that this
should be read Wistlan wudu is therefore impossible. Besides, the stroke over
the word is never used in the Exeter Book for n, which is always written in
full, except in the word \>on (\>onne), and in Latin words ; the stroke is fre
quently used, but in all other cases signifies m.
123. Rcedhere and Rondhere. Miillenhoff (Z.f.d.A. vi, 453) would connect
them with the Randolt unde Rienolt who are mentioned in Biterolf, together
with the Harlungs and others, as retainers of Ermanaric (5723, 4604 etc.),
" denn die namen bedeuten ganz dasselbe." Possibly Bondhere is the sarne
hero as the Bandver who in the Scandinavian versions is Ermanaric's son, the
equivalent of the Frederic of the southern versions. The second element in
proper names is particularly apt to be corrupted. See Introduction, p. 23.
Rumstan. Almost certainly, as was pointed out by Lappenberg (181) and
subsequently by Hertz (Deutsche Sage im Elsass, Stuttgart, 1872, p. 220), the
Bimstein who is mentioned in Biterolf as a retainer of Ermanaric, and more
particularly of the two young Harlungs :
— Von den Harlungen
Fritelen dem jungen
unde ouch Imbrecken,
den volgeten die recken
Wahsmuot unde Rimstein,
ez ween der tac nie beschein
bezzer wigande. Biterolf, 10673-9 (cf. 4771)-
The name also occurs in the Thidreks Saga, 147-149, where Bimstein is a
tributary earl of jErminric but rebels against him. Thithrek gives help to
.iErminric, and Vithga (Wudga) slays Bimstein in a sortie.
Bassmann in his translation of the Thidreks Saga (Die deutsche Heldensage
und ihre Heimat, n, 459, 1863) also pointed out that Bimstein is probably
identical with our Bumstan. See also Grimm, Heldensage, 3" Aufl., Anhang,
462 ; Jiriczek(2) 79.
Gislhere. See Introduction, p. 65.
124. Wi\>ergield. " Withergield ist dem Beowulfsliede angehorig," Lappen
berg (181). If the Withergyld of Beowulf (2051) is the name of a hero (as it is
taken, e.g. by Gering and Schiicking), and not a common noun, he would be a
leading figure in the combat between Danes and Heathobeardan, presumably a
Heathobeard.
Fre.o\eric. Probably the Friderich who is frequently mentioned in Dietrich's
Flucht, son of Ermanaric, who is at last destroyed by his father. (See Intro
duction, p. 30, note. ) There is however another Friderich, a retainer of Dietrich
(Dietrich's Flucht, Rabenschlacht). The historic prototype of one or both of
these figures is probably to be found in Frederic, king of the Bugians, a
picturesque scoundrel of the later fifth century, who first joined and then
deserted Theodoric. (See Hodgkin's Italy, m, 164-209 ; and cf. Matthaei in
Text (tt. 122-139) 223
J?eahJ?e ic hy anihst nemnan sceolde.
Ful oft of bam heape hwinende fleag
giellende gar on grome J?eode :
wrseccan ]?ser weoldan wundnan golde,
130 werum ond wifum, Wudga ond Hama.
Swa ic ]>cet symle onfond || on bgere feringe,
bset se bib leofast londbuendum,
se ]>e him god syle'5 gumena rice
to gehealdenne, benden he her leofaS." —
135 Swa scribende gesceapum hweorfao"
gleomen guraena geond grunda fela,
bearfe secgaS, boncword sprecab,
simle suS o)?be nortJ sumne gemetaS
gydda gleawne, geofum unhneawne,
last. Full oft from that company flew the spear, whistling and shrieking,
against the hostile folk. Wudga and Hama, wanderers o'er the earth,
ruled there, by wounden gold, over men and women.
So have I ever found it in my journeying, that he is most beloved i
to the dwellers in the land, to whom God giveth dominion over men to
hold it whilst he liveth here."
So are the singers of men destined to go wandering throughout many
lands : they tell their need, they speak the word of thanks : south or north
they ever meet with one, skilled in songs, bounteous in gifts, who desires
Z.f.d.A. XLIII, 326). Jiriczek(2) 73 regards the identity of the Freotheric
mentioned here with the son of Ermanaric as certain. But it is hardly so :
cf. Panzer, H.i.B. 78. Yet I think that the balance of probability is in favour
of all the names being connected: and that Widsith records a stage of the
tradition, at which the king Frederic of history has come to be regarded as a
retainer of Ermanaric, but not yet as his son.
126. ic hy anihst. ic hivan nyhst, Leo, Ettmiiller (1). "Obwohl ich's
Hausvolk zunachst nennen musste " (Ettmiiller). Holt, reads aneist, for
metrical reasons. See Introduction, p. 173.
127. hwinende. ctVa£ \ey6/j,et>oi>. Cf. O.N. hwina.
128. giellende gar. Bugge sees an imitation of this line in Atlakvfya 15 :
me\> geire gjallanda at vekja gram hilde. See P.B.B. xxxv, 245, and H. D. tr.
Schofield xxiii ; cf. Neckel, 468, note 4.
129. \<KV. Where? We seem, just as in 11. 50, 88, to have a reference to
something not explained in the extant poem. This, with the allusions to
Ealhhild and Scilling, favours the view that either Widsith has been muti
lated, or that the background of legend was so familiar as to render the most
obscure references intelligible. J>#r may perhaps refer to the stronghold of
Hama, and may be identical with the byrhtan byrig to which in Beowulf he
carries off the Brosinga mene.
wundnan golde. Leo, Ettmuller(i) took wundnan golde as qualifying Wudga
and Hama "die Wackern walteten da, bewunden mit Golde." So Thorpe,
Cod. Ex. But wundnan golde is instrumental ; see Introduction, pp. 167,
170.
wundnan was altered by Kernble (2) to wundum [for loundnum] ; wundnum,
Ettmiiller d) (note, not text), Ettmiiller (2) text.
224 Widsith
140 se Ipe fore dugu)?e wile dom araeran,
eorlscipe sefnan, o)> )>aet eal scseceS,
leoht ond lif somod : lof se gewyrceS,
hafaS under heofonura heahfaestne dom.
to exalt his fame before his chieftains — to do deeds of honour: till all
departeth, life and light together : he gaineth glory, and hath, under the
heavens, an honour which passeth not away.
131-4. Miillenhoff condemns these four lines as an interpolation on account
of their silliness (Z.f.d.A. xi, 293, 4).
symle onfond. onfond symle, Holt, for metrical reasons.
135-143. Widsith's discourse is ended, and these nine lines form an
epilogue, similar to the nine lines of prologue prefixed to it. Compare the way
in which the Wanderer's discourse is fitted with a Prologue and an Epilogue.
These in each case may well be later additions, but evidence is not as strong
against these closing lines as against 11. 1-9. These verses are somewhat
commonplace, but not faulty, or out of harmony with the rest of the poem.
Cf. Introduction, p. 151.
Miillenhoff finds them "voll Schwung und Erhabenheit" (Z.f.d.A. xi,
293). Moller on the other hand finds in them "einfachste niichternheit "
(V.E. 36).
140. fore dugu\>e, before the assembled company. Cf. Juliana 256 ;
Beowulf, 2020.
APPENDIX.
(A) BIBLIOGRAPHY.
(1) PERIODICALS AND COLLECTIONS
Aarbfyg&r f. nord. Oldk. Aarb0ger for Nordisk Oldkyndighed og Historic.
Kj0benhavn, 1866, etc.
Anglia. Anglia, Zeitschrift fxir Englische Philologie, herausgegeben von
R. P. Wiilcker [Eugen Einenkel]. Halle, 1878, etc.
A.f.d.A. Anzeiger fiir deutsches Alterthum. Berlin, 1876, etc. (Z.f.d.A.
xix = A.f.d.A. I.)
A.f.n.F. Arkiv for nordisk Filologi udgivet ved G. Storm [A. Kock].
Christiania, 1883, etc.
A.T.f.S. Antiqvarisk Tidskrift for Sverige. Stockholm, 1863, etc.
(Herrig's) Archiv. Archiv fur das Studium der neueren Sprachen und
Litteraturen. Elberfeld, 1846, etc.
Quoted according to the original numbering.
D.L.Z. Deutsche Literaturzeitung. Berlin, 1880, etc.
E.E.T.S. Early English Text Society.
Engl. Stud. Englische Studien, herausgegeben v. Eugen Kolbing. Heil-
bronn, 1877, etc.
Eriu. Eriu : The Journal of the School of Irish Learning, Dublin. Dublin,
1905, etc.
Folk Lore : a quarterly review. London, 1890, etc.
Oermania. Germania, Vierteljahrsschrift fiir deutsche Alterthumskunde.
Wien, 1856-92.
I.F. Indogermanische Forschungen, herausgegeben von K. Brugmann u.
W. Streitberg. Strassburg, 1891, etc.
J.d. V.f.n.S. Jahrbuch des Vereins fiir" niederdeutsche Sprachforschung.
Norden u. Leipzig, 1876, etc.
Langebek. Scriptores rerum Danicarum, collegit J. Langebek. Hafniae,
1772-1878.
Litteraturblatt f. Philol. Litteraturblatt fiir germanische u. romanische
Philologie herausg. v. O. Behagel u. F. Neuman. Heilbronn, Leipzig,
1880, etc.
c. 15
226 Widsith
Mod. Philol. Modern Philology. Chicago. University of Chicago Press,
1903, etc.
M.O.H. Monumenta Germaniae Historica, edidit Societas aperiendis fon-
tibus rerum Germanicarum medii aevi. 4° Berolini apud Weidmannos,
1877, etc. ; 4° Hannoverae, 1879, etc.
M.L.N. Modern Language Notes. Baltimore, 1886, etc.
M.L.R. The Modern Language Review, ed. by John G. Robertson.
Cambridge, 1906, etc.
Morsbachs Studien. Studien zur englischen Philologie. Halle, 1897, etc.
Nordalb. Stud. Nordalbingische Studien, 6 Bde. Kiel, 1844-54.
P.B.B. Beitrage zur Geschichte der deutschen Sprache und Litteratur,
herausgeg. v. Hermann Paul u. Wilhelm Braune. Halle a. S., 1874, etc.
Pauls Ordr. d). Grundriss der germanischen Philologie, herausgegeben
von Hermann Paul. Strassburg, 1889, etc.
Pauls Ordr. (2). [The same], zweite Auflage. Strassburg, 1896, etc.
Pertz (fol.). Monumenta Germaniae Historica, edidit G. H. Pertz. Fol.
Hannoverae, 1826, etc.
Q.F. Quellen u. Forschungen zur Sprach- und Culturgeschichte der ger
manischen Volker, herausgegeben von B. ten Brink und W. Scherer.
Strassburg, 1874, etc.
W.S.B. Sitzungsberichte der k. Akad. der Wissenschaften zu Wien.
Phil. -hist. Classe. Wien, 1850, etc.
Z.f.d.A. Zeitschrift fur deutsches Alterthum [und deutsche Literatur],
herausgegeben von Moriz Haupt [K. Mullenhoff, E. Steinmeyer, etc.].
Leipzig, Berlin, 1841, etc.
(Zacher's) Z.f.d.Ph. Zeitschrift fur deutsche Philologie, begriindet von
Julius Zacher. Halle, 1869, etc.
(2) EDITIONS OF WIDSITH (IN CHRONOLOGICAL ORDER).
CONYBEARE, Illustrations. Illustrations of Anglo-Saxon poetry, by J. J.
Conybeare, ed. by his brother, W. D. Conybeare. London, 1826.
pp. 10-27, Text, Notes, Latin Translation and English paraphrase.
KEMBLE, Beowulf. The Anglo-Saxon Poems of Beowulf, the Traveller's
Song and the Battle of Finnes-Burh, ed. by John M. Kemble.
London, 1833. pp. 221-233, Text.
Kemblds Text is reproduced in Latham's Germania of Tacitus,
London, 1851. KemblJs second edition (1835) differs considerably
from his first.
GUEST, Rhythms. A History of English Rhythms, by Edwin Guest.
2 vols. London, 1838. pp. 78-93, Text and English Translation.
LEO, Sprachproben. Altsachsische und Angelsachsische Sprachproben,
herausgeg. und mit einem erklarenden Verzeichniss der angelsachs-
ischen Worter versehen von Heinr. Leo. Halle, 1838. pp. 75-85,
Text, Translation and Notes.
Appendix 227
ETTMULLER (t). Scopes vldsidh. Sangers Weitfahrt. ^Edhelstans Sieg
bei Brunanburg. Angelsachsisch und deutsch von L. Ettmiiller.
Zurich, 1839.
THORPE, Cod. Ex. Codex Exoniensis. A collection of Anglo-Saxon
Poetry, ed. by Benj. Thorpe. London, 1842. pp. 318, etc.
Thorpe's Translation is reproduced in Latham's Oermania of
Tacitus, 1851.
EBELING. Angelsachsisches Lesebuch von F. W. Ebeling. Leipzig, 1847.
pp. 97-101.
SCHALDEMOSE. Beo-Wulf og Scopes WidsitS, to angelsaxiske digte, med
overssettelse og oplysende anmserkninger udgivne af F. Schaldemose.
Kj^benhavn, 1847, second edit. 1851. pp. 176-188.
KLIPSTEIN. Analecta Anglosaxonica. Selections, in prose and verse,
from the Anglo-Saxon Literature, by Louis F. Klipstein. 2 vols.
New York, 1849. Vol. n : Text, 299-307 ; Notes, 422-430.
ETTMULLER (2). Engla and Seaxna Scopas and B&ceras. Anglo-Saxonum
poetae et scriptores prosaici, ed. L. Ettmullems. Quedlinburgii et
Lipsiae, 1850. pp. 208-211.
THORPE, Beowulf. The Anglo-Saxon Poems of Beowulf, the Scop or
Gleeman's Tale and the Fight at Finnesburg. With a literal transla
tion, notes, glossary, etc., by Benj. Thorpe. Oxford, 1855. pp. 215-227.
MULLER, Angelsachsisches Lesebuch. c. 1855. 22. Des Sangers Reisen.
/ have not seen this edition, which was privately printed. There is
no copy in the British Museum, nor, apparently, in any of the German
University libraries.
GREIN, Bibliothek. Bibliothek der angelsachsischen Poesie, in kritisch
bearbeiteten Texten und mit vollstandigen Glossar herausgeg. von
C. W. M. Grein. 4 Bde. Gottingen, 1857. pp. 251-255.
RIEGER, Lesebuch. Alt- und angelsachsisches Lesebuch nebst altfriesischen
Stucken mit einem Wb'rterbuche von Max Rieger. Giessen, 1861.
pp. 57-61.
WULCKER, Kleinere Dichtungen. Kleinere angelsachsische Dichtungen,
versehen von R. P. Wulcker. Halle, 1882. pp. 1-6.
MOLLER, V. E. Das altenglische Volksepos in der urspriinglichen stro-
phischen Form, von Hermann Moller. Kiel, 1883. pp. 1-39, i-vi.
GREIN- WULCKER. Bibliothek der angelsachsischen Poesie, begriindet
von C. W. M. Grein, neu bearbeitet von R. P. Wulcker. 3 Bde.
Cassel, 1883, etc. Vol. I, pp. 1-6.
KLUGE, Lesebuch. Angelsachsisches Lesebuch, zusammengestellt von
Friedrich Kluge. 2te Auflage. Halle, 1897. pp. 123-126.
HOLT. Beowulf, nebst den kleineren Denkmalern der Heldensage,
herausgeg. von F. Holthausen. 2te Auflage. Heidelberg, 1908-9.
Text of Widsith in vol. I : Bibliography and notes in vol. n.
SEDG. Beowulf, ed. by W. J. Sedgefield. Manchester, 1910.
Text of Widsith, pp. 139-142.
15—2
228 Widsith
Comments or translations without text will be found in
LAPPENBERG. [A review of Leo's Sprachproben appeared in the] Jahrbiicher
fur wissenschaftliche Kritik, August, pp. 169-182, by J. M. Lappen-
berg. Berlin, 1838, n.
MiiLLENHOFF in Nordalb. Stud. Die deutschen Volker in Nord- und
Ostsee in altester Zeit. In Nordalbingische Studien, Kiel, 1844,
I, 111-174.
MULLENHOFF in Z.f.d.A. Zur Kritik des angelsachsischen Volksepos. 2.
Widsith. In Z.f.d.A., xi, 275-294.
SCHIPPEE in Oermania. Zum Codex Exoniensis von J. Schipper. In
Germania, xix, 327-339 (especially 333).
GUMMERE d). Widsith, a translation, with brief commentary, in Modern
Language Notes, iv, 419 etc. (1889).
MORLEY'S English Writers contains a translation into blank verse : re
printed in Cook and Tinker's Select Translations, Boston, 1902.
VOGT, Beowulf. Beowulf, iibersetzt von Paul Vogt. Halle, 1905.
pp. 100-103 give a useful abstract of Widsith, with commentary.
LAWRENCE in Mod. Philol. Structure and interpretation of Widsith, by
W. W. Lawrence. In Modern Philology, iv, 1906, pp. 329-374.
SCHUTTE, Oldsagn, 1907, contains a translation of Widsith into Danish.
GUMMERE (2). The Oldest English Epic. Beowulf... Widsith, translated
by F. B. Gummere. New York, 1909.
SLEBS. WfdsfS. In Festschrift Wilhelm Vietor. Marburg, 1910.
ANSCOMBE. Widsith. In Anglia, xxxiv, pp. 526-7.
(3) Other O.E. poems have been quoted from the edition of Grein-Wulcker.
Holder's Beowulf (1899) has also been used. Editions of other works
frequently quoted are :
ABELING. Das Nibelungenlied und seine Literatur. Leipzig, 1907.
(Teutonia 7.)
AGATHIAS. Agathiae Historiarum libri quinque ; B. G. Niebuhrius
recensuit. Bonnae, 1828. In the Corpus Scriptorum Historiae
Byzantinae.
ANDERSON. The Anglo-Saxon Scop, 1903. (Univ. of Toronto Studies.)
Antiquite's Russes. Antiquite's Russes [edited by C. C. Eafn, etc.']. Copen
hagen, 1850. Published by the Kongeligt Nordisk Oldskrift-Selskab.
BARNOUW. Textcritische Untersuchungen. Leiden, 1902.
BEDE, H.E., ed. Holder. Bedae Historia Ecclesiastica, ed. Holder.
Freiburg, 1882. Holder's Germanische Bucherschatz, 7.
BEDE, H.E., ed. Plummer. Venerabilis Bedae Historiam Ecclesiasticam
...Historiam Abbatum...recog. C. Plummer. 2 torn. Oxonii, 1896.
BINDING. Die burgundisch-romanische Konigreich, von C. Binding.
Leipzig, 1868.
Appendix 229
BINZ in P.B.B. Zeugnisse zur germanischen Sage in England, von Q.
Binz. In P.B.B. xx, 141-223 (1895).
BIRCH, Cart. Sax. Cartularium Saxonicum, a collection of charters
relating to Anglo-Saxon history, by W. de Gray Birch. 3 vols.
London, 1885-93.
BLUHME. Die Gens Langobardorum und ihre Herkunft. Bonn, 1868.
BOER, Ermanarich. Die Sagen von Ermanarich und Dietrich von Bern
von R. C. Boer. Halle, 1910. Zachers Germanistische Hand-
bibliothek, x.
BOTTGER. Diocesan u. Gaugrenzen Norddeutschlands v. H. Bb'ttger.
4 Bde. Halle, 1875-6.
IV, 214-17, 224-9 particularly helpful.
BRANDL in Pauls Ordr. (2). Geschichte der altenglischen Literatur von
Alois Brandl. Strassburg, 1908. In Pauls Grundriss, n, 1.
BRANDL in Archiv. Zur Gotensage bei den Angelsachsen. In Herrigs
Archiv, cxx (1908), pp. 1-8.
BREMER in Pauls Grdr. (2). Ethnographie der germanischen Stamme von
Otto Bremer. In Pauls Grundriss, in, 735-950 (Abschnitt xv).
This section is to be found in the second edition only.
BUGGE, H.D. Helge-Digtene i den aeldre Edda : deres hjem og forbindelser,
af Sophus Bugge. Kj0benhavn, 1896.
BUGGE tr. Schofield. [A revised translation of the above.] London,
1899.
C.P.B. Corpus poeticum boreale : the poetry of the old Northern tongue,
edited by Gudbrand Vigfusson and F. York Powell. 2 vols. Oxford
1883.
CASSIODORUS, Variae. Cassiodori Senatoris Variae, recensuit Th
Mommsen. Berolini, 1894. In the Monumenta Germaniae Historica,
Auct. Ant. xii.
CHAD WICK. The origin of the English Nation. Cambridge, 1907.
CHADWICK in Camb. Hist. " Early National Poetry " in the Cambridge
History of English Literature, vol. I, chap, ill (pp. 19-40).
Chronicle, ed. Plummer. Two of the Saxon Chronicles parallel, edited by
Charles Plummer on the basis of an edition by John Earle. 2 vols.
Oxford, 1892-99. See particularly notes on the Genealogies, n,
pp. 1-6.
CLARK. Sidelights on Teutonic History during the Migration Period.
Cambridge, 1911.
DAHLMANN. Forschungen auf dem Gebiete der Geschichte. Altona,
1822.
Danmarks Riges Historic. Danmarks Riges Historic af Joh. Steenstrup,
etc., I. Kj0benhavn, 1897-1904.
Edda, ed. Symons and Gering. Die Lieder der Edda, herausgegeben von
B. Symons und H. Gering. 2 Bde. Halle a. S., 1888, etc. Zachers
Germanistische Handbibliothek, vn.
230 Widsifh
Edda, ed. Better and Heinzel. Saemundar Edda, mit einem Anhang
herausgegeben und erklart von F. Detter u. R. Heinzel. 2 Bde.
Leipzig, 1903.
ERDMANN in A. T.f.S. Om folknamnen Gotar och Goter, af Axel Erdmann.
In A.T.f.S., xi, 4.
ERDMANN. tfber die Heimat und den Namen der Angeln von A. Erdmann
in the Skrifter utgifna af Humanistiska Vetenskapssamfundet i
Upsala, I, 1890.
FAHLBECK in A.T.f.S. Beovulfsqvadet sasom kalla for nordisk forn-
historia af Pontus Fahlbeck. In A.T.f.S., vm, 2.
Fornaldar Sqgur, ed. Rafn. Fornaldar SQgur Nordrlanda eptir gginlum
Handritum utgefnar af C. C. Rafn. Kaupmannaho.fn, 1829. Bd I
contains : pp. 1-109, Saga Hrolfs Konungs Kraka ; pp. 113-
234, Vglsunga Saga ; pp. 409-533, Hervarar Saga ok Heidreks
Konungs.
FORBTEMANN. Altdeutsches Namenbuch. 2 Bde. 2te Auflage. Bonn,
1900.
GERING. Beowulf, ubersetzt u. erlautert von H. Gering. Heidelberg,
1906.
Germania, ed. Church and Brodribb. The Germania of Tacitus, with a
revised text, English notes, and map, by A. J. Church and W. J.
Brodribb. London, 1903.
Germania, ed. Holder. Cornelii Taciti de origine et situ Germanorum
liber, ed. A. Holder. Freiburg, 1882. In Holder's Germanischer
Biicherschatz.
Germania, ed. Latham. The Germania of Tacitus, with ethnological
dissertations and notes. London, 1851.
GIBBON, ed. Bury. The History of the Decline and Fall of the Roman
Empire, edited by J. B. Bury. 7 vols. London, 1896-1900.
GOLTHER. Handbuch der germanischen Mythologie. Leipzig, 1895.
GREGORII Hist. Franc. Historia Francorum, ed. Arndt. Hannoverae,
1885. In the Monumenta Germaniae Historica : Scriptores rerum
Merovingicarum, i; Gregorii Turonensis opera.
GRIMM, Heldensage. Die deutsche Heldensage von Wilhelm Grimm.
Gottingen, 1829. Second edition, revised by Miillenhoff, Berlin, 1867 ;
third edition, a reprint of the first, revised by R. Steig, Gutersloh,
1889.
References to the first edition, the pagination of which is noted in
the margin of the third.
GRIMM, D.M. Deutsche Mythologie, von Jacob Grimm. Gottingen,
1835. Second edit., Gottingen, 1844 ; third, Gottingen, 1854 ; fourth,
revised by Meyer, 3 vols., Weimar, 1875. Teutonic Mythology,
translated from the fourth edition, with notes and appendix by
J. S. Stallybrass. 4 vols. London, 1879.
References to the first edition, and to Stallybrass' translation.
Appendix 231
GRIMM, G.D.S. Qeschichte der deutschen Sprache, von Jacob Grimm.
Leipzig, 1848. Second edit., 1853 ; third, revised by Mullenhoff, 1867 ;
fourth, 1880.
References to the first edition, the pagination of which is noted in
the margin of the fourth.
GRIMM, R.A. Deutsche Rechtsalterthiimer. Gottingen, 1828. Fourth
edition, Leipzig, 1899.
References to the first edition, the pagination of which is noted in
the margin of the fourth.
HAIGH, Sagas. The Anglo-Saxon Sagas ; an examination of their value as
aids to history, by Daniel H. Haigh. London, 1861.
Heimskringla af Snorri Sturluson, udg. ved F. Jonsson. 4 vols. K0ben-
havn, 1893-1901.
HEINZEL, Hervararsaga in W.S.B. Ueber die Hervararsaga, von Richard
Heinzel. In the Sitzungsberichte der philosophisch-historischen Classe
der k. Akademie der Wissenschaften zu Wien, Bd cxiv, 417 (1887).
HEINZEL, Ostgotische Heldensage in W.S.B. Ueber die ostgotische
Heldensage, in the same, Bd cxix, Abth. 3.
HEINZEL in A.f.d.A. Das altenglische Volksepos, von Hermann Moller.
[A review by Richard Heinzel] in the Anzeiger fiir deutsches
Alterthum x, 215-33.
Heldenbuch. Deutsches Heldenbuch. 5 Bde. Berlin, 1866-70.
HEUSLER, L.u.E. Lied und Epos in germanischer Sagendichtung, von
Andreas Heusler. Dortmund, 1905.
HEYNE-SCHUCKING. Beowulf herausgeg. v. M. Heyne. Neunte Auflage
bearbeitet von L. L. Schiicking. Paderborn, 1910.
Historia Brittonum, ed. Stevenson. Nennii Historia Britonum, recensuit
Josephus Stevenson. Londini, 1838.
Historia Brittonum, ed. Mommsen in N.G.H. Historia Brittonum cum
additamentis Nennii, in Mommsen's Chronica Minora saec. iv, v, vi,
vn. Berolini, 1898 (Mon. Ger. Hist.).
HODGKIN, Italy. Italy and her Invaders, by Thomas Hodgkin. Second
edit. 7 vols. Oxford, 1892, etc.
HOOPS, Waldbaume. Waldbaume und Kulturpflanzen im germanischen
Altertum, von Johannes Hoops. Strassburg, 1905.
Hoops Reallexikon. Reallexikon der germanischen Altertumskunde.
Strassburg, 1911.
Bd. I, Lief. 1 (A — BA) so far published.
JAHN. Die Geschichte der Burgundionen. 2 Bde. Halle, 1874.
JIRICZEK d). Die deutsche Heldensage, von 0. L. Jiriczek. 3te Auflage.
Leipzig, 1906. Sammlung Goschen, 32.
JIRICZEK (2). Die deutsche Heldensage, von 0. L. Jiriczek. Vol. i.
Strassburg, 1898.
Vol. I deals with the stories of Weland, Ermanaric, and Theodoric
of Verona.
232 Widsith
JORDAN. Die Heimat der Angelsachsen, in the Verhandlungen der 49
Versammlung deutscher Philologen, Basel, 1907, pp. 138-140.
JORDANES, ed. Holder. lordanis de origine actibusque Getarum, ed.
Holder. Freiburg, 1882. In Holder's Germanischer Biicher-
schatz, 5.
JORDANES, ed. Mommsen. lordanis Romana et Getica, ed. Theodoras
Mommsen. Berolini, 1882. In the Monumenta Germaniae Historica,
Auct. Ant. v, 1.
Unless otherwise stated references are to the Getica.
KEMBLE'S Saxons. The Saxons in England, by J. M. Kemble. 2 vols.
London, 1849.
KER, Epic and Romance. London, 1896.
KER, The Dark Ages. Edinburgh, 1904.
KOCK, Ynglingar. Om Ynglingar sasom namn pa en svensk konungaatt
in the Historisk Tidskrift (Stockholm), 1895.
KOEQEL, Ltg. Geschichte der deutschen Litteratur bis zum Ausgange
des Mittelalters, von Rudolf Koegel, 1, 1, 1894 ; i, 2, 1897. Strassburg.
Particularly 1,96-175; n, 191-219.
Kudrun, ed. Martin (2). Kudrun, herausgegeben und erklart von Ernst
Martin. Zweite Auflage. Halle, 1902. Zachers Gennanistische
Handbibliothek.
LAPPENBERG, Geschichte. Geschichte von England, von J. M. Lappenberg.
2 Bde. Hamburg, 1834.
LAPPENBERG, trans. Thorpe. History of England under the Anglo-
Saxon kings, translated from the German of Dr J. M. Lappenberg
by Benjamin Thorpe, with additions... by the author. London,
1844-5.
MEYER. Altgermanische Religionsgeschichte von Richard M. Meyer.
Leipzig, 1910.
MOGK in Pauls Grdr. (2). Mythologie, in Grundriss, in, 230-406.
MORLEY'S English Writers. English Writers. An attempt towards a
history of English Literature, by Henry Morley. 14 vols. London,
1887-95. Particularly vol. n, pp. 1-40.
MUCH \\\' P.B.B. "Die Siidmark der Germanen," "Die Germanen am
Niederrhein," " Goten und Ingvaeonen," von Rudolf Much. In Paul
u. Braunes Beitrage, xvn, 1-225 (1893).
MUCH. Deutsche Stammeskunde, von R. Much. 2te Aufl. Leipzig,
1905.
MULLENHOFF, Beovulf. Beovulf. Untersuchungen iiber das angelsachs-
ische Epos und die alteste Geschichte der germanischen Seevolker,
von Karl Miillenhoff. Berlin, 1889.
MULLENHOFF, D.A.K. d). Deutsche Altertumskunde, von Karl Miillen
hoff. Berlin, 1870, etc.
MULLENHOFF, D.A.K. (2). [Second edition, revised by Max Roediger.]
Berlin, 1890, etc.
Appendix 233
MULLENHOFF in Nordalb. Stud. Die deutschen Volker an Nord- und
Ostsee in altester Zeit. In Nordalbingische Studien, i (1844), 111.
MULLENHOFF, Z.E., or MULLENHOFF in Z.f.d.A, xn. Zeugnisse und
Excurse zur deutschen Heldensage von Karl Miillenhoff. In the
Z.f.d.A. xii, 253 etc., 413 etc. Subsequent additions by 0. Jaenicke,
Z.f.d.A. xv, 310 etc.
MULLER. Mythologie der deutschen Heldensage, von W. Miiller.
Heilbronn, 1886.
NECKEL. Beitrage zur Eddaforschung. Dortmund, 1908.
Nibelungen Lied. Der Nibelunge N6t, herausg. von Karl Bartsch. 2 Bde.
Leipzig, 1870-6.
OLRIK, Heltedigtning. Danmarks Heltedigtning, en Oldtidsstudie af Axel
Olrik, i Rolfkrake og den seldre Skjoldungraekke, II Starkad den
Gamle. K0benhavn, 1903, 1910.
OLRIK, Sakses Oldhistorie. Sakses Oldhistorie, af Axel Olrik. Kj0benhavn,
1892-4.
Pt 1 appeared in the Aarbfyger for Nor disk Oldkyndighed, n, 7, 1 ;
Pt 2 separately.
PANZER, H.G. Hilde-Gudrun. Eine sagen- und literargeschichtliche
Untersuchung. Halle, 1901.
PANZER, H.i.B. Deutsche Heldensage im Breisgau. Heidelberg, 1904.
PANZER, Beowulf. Studien zur germanischen Sagengeschichte von F.
Panzer, i Beowulf. Munchen, 1910.
PAULUS. Pauli Historia Langobardorum, ed. L. Bethmann et G. Waitz.
Hannoverae, 1878. In the Monumenta Germaniae Historica — Scrip-
tores rerum Langobardicarum et Italicarum saec. vi-ix.
PLINY, Nat. Hist. C. Plini Secundi naturalis historiae libri xxxvn post
Ludovici obitum edidit C. Mayhoff. Lipsiae (Teubner), 1906.
PROCOPIUS. Procopius ex recensione G. Dindorfii. 3 torn. Bonnae, 1833.
In the Corpus Scriptorum Historiae Byzantinae.
PTOLEMY, ed. Miiller. Cl. Ptolemaei geographia, recognovit C. Miillerus.
Vol. i. Parisiis, 1883 (Tabulae, 1901).
RAJNA. Le Origini dell' Epopea francese. Firenze, 1884.
RASSMANN. Die deutsche Heldensage. 2te Ausgabe. Hannover, 1863.
Contains a German translation of the Thidreks Saga.
RICHTER. Chronologische Studien zur angelsachsischen Literatur. Halle,
1910. (Morsbachs Studien, xxxin.)
RIEGER in Z.f.d.A, Ingavonen, Istavonen, Herminonen von Max Rieger.
In the Z.f.d.A. xi, 177-205.
ROEDIGER. Die Sage von Ermenrich u. Schwanhild. In Weinhold's
Zeitschrift des Vereins fiir Volkskunde, I, 1891, pp. 242 etc.
SARRAZIN in Engl. Stud. Neue Beowulfstudien, xxm, 221-67 ; xxxv,
19 etc. ; XLII, 1-37. See also xxiv, 144-5.
SAXO, ed. Holder. Saxonis grammatici Gesta Danorum, herausgegeben
von Alfred Holder. Strassburg, 1886.
234 Widsith
SAXO, tr. Elton. The first nine books of the Danish History of Saxo
Grammaticus, translated by Oliver Elton. London, 1894. Publica
tions of the Folk Lore Society, xxxni.
SCHUCK. Folknamnet Geatas i den fornengelska dikten Beowulf. Upsala,
1907.
SCHUTTE in A.f.N.F. Gudmund Schiitte — Anganty-Kvadets Geografi.
In the Arkiv for Nordisk Filologi, xxi, 30-44.
SCHUTTE, Oldsagn. Oldsagn om Godtjod, med saerligt henblik pa Folke-
stamsagn. Kj0benhavn, 1907.
SEARLE. Onomasticon Anglosaxonicum. Cambridge, 1897.
SEEBOHM. Tribal Custom in Anglo-Saxon Law. London, 1902.
SEELMANN in J.d. V.f.n.S. Abhandlungen zur Geschichte der deutschen
Volksstamme Norddeutschlands und Danemarks. In the Jahr-
buch des Vereins fiir niederdeutsche Sprachforschung, xu, 1886,
pp. 1-93.
SNORRA Edda. Snorri Sturluson, Edda udgiven af Finnur J6nsson.
K0benhavn, 1900.
STOPFORD BROOKE, Early Eng. Lit. The History of Early English Litera
ture, by Stopford A. Brooke. 2 vols. London, 1892. Particularly
Vol. I, Chapters i and vi, and Appendices A and B.
SYMONS in Pauls Ordr. (2). Heldensage, von B. Symons. In Pauls
Grundriss, vol. in, pp. 606-734.
TEN BRINK, Beowulf. Beowulf. Untersuchungen von Bernhard ten
Brink. Strassburg, 1888. (Q.F. 62.)
A review of Ten Brink's Beowulf, by H. Holler, will be found in
Englische Studien, xni, 247.
TEN BRINK, in Pauls Ordr. d). Altenglische Litteratur. Particularly
pp. 510-608 of vol. n, pt. 1 of the first edition.
Thidreks Saga. Saga DiSriks konungs af Bern, i Norsk bearbeidelse fra
det trettende Aarhundrede efter Tydske Kilder. Udgivet af C. R.
Unger. Christiania, 1853.
Thidreks Saga. piSriks Saga af Bern, udgivet ved H. Bertelsen. K0ben-
havn, 1905, etc.
References to Unger's edition. Unger's numbering is given (in
brackets') in Bertelsen.
UHLAND, Schriften. Schriften zur Geschichte der Dichtung und Sage, von
Ludwig Uhland. 8 Bde. Stuttgart, 1865-73.
Vglsunga Saga, ed. Wilken. Die prosaische Edda, im Auszuge, nebst
Vglsunga Saga und Nornagests-thattr, herausgegeben von Ernst
Wilken. Paderborn, 1877-83.
WAITZ. Forschungen zur deutschen Geschichte. I. Der Kampf der
Burgunder u. Hunen. Gottingen, 1862.
WEILAND. Die Angeln. Ein Capitel aus der deutschen Alterthumskunde
von L. Weiland. Sonderabdruck aus der Festgabe fur Georg Hanssen.
Tubingen, 1889.
Appendix 235
WEYHE in EngL Stvd. Konig Ongentheows Fall. In Engl. Stud, xxxix,
14-39.
WULKER, Ordr. Grundriss zxir Geschichte der Angelsachsischen Lit-
teratur. Leipzig, 1885. Particularly pp. 244-346: Angelsachsische
Heldendichtung.
ZEUSS. Die Deutschen und die Nachbarstamme, von K. Zeuss
Miinchen, 1837.
(B) "MAURUNGANI" AND THE "GEOGRAPHER OF
RAVENNA."
The " geographer " divides the points of the compass into twenty-four
hours, twelve northern (of the night), and twelve southern (of the day).
Following this arrangement, we have first the land of the Germans
(Franks) with Britain beyond : then the land of the Frisians, then of
the Saxons, then of the Northmen or Danes, with the Elbe country in
front (of old called Maurungani), Then come the Scridefinni, then Scythia.
We have thus travelled through six hours, i.e. ninety degrees. The text
runs :
Prima ut hora noctis Germanorum est patria quse modo a Francis
dominatur. ..cujus post terga infra Oceanum praedicta insula Britania....
Secunda ut hora noctis ex parte ipsa Germania vel FrixonumDorostates
est patria... Tertia ut hora noctis Saxonum est patria... Quartaut hora
noctis Northomanorum est patria, quse et Dania ab antiquis dicitur,
cujus ad frontem Alpes [read Albes] vel patria Albis : Maurungani
certissime antiquitus dicebatur, in qua Albis patria per multos annos
Francorum linea remorata est, et ad frontem ejusdem Albis Datia
minor dicitur, et dehinc super ex latere magna et spatiosa Datia
dicitur, quse modo Gipidia ascribuntur, in qua nunc Unorum gens
habitare dinoscitur....Quinta ut hora noctis Scirdifrinorum vel
Rerefenorum est patria. Sexta ut hora noctis Scytharum est
patria, unde Sclavinorum exorta est prosapia1.
The second mention of [? Maur~]ungani comes in Chapter iv, § 18,
between the description of Saxony (§ 17) and Pannonia (§ 19).
Item ad partem quasi meridianam, quomodo a spatiosissima
dicatur terra, est patria quse dicitur Albis [ Jungani, montuosa
per longum, quasi ad orientem multum extenditur, cujus aliqua pars
Baias dicitur.... Haec patria habet non modica flumina, inter cetera
fluvius grandis qui dicitur Albis et Bisigibilias sexaginta quse in
Oceano funduntur.
Baias is apparently Bohemia. For the unintelligible Bisigibttias
1 Ravennatis anonymi cosmographia, ed. Finder and Parthey, Berolini, 1860,
pp. 27, 28 (§ i, 11).
236 Widsiih
sexaginta it has been proposed to read Visurgi et alia sexaginta.
Miillenhoff supposes it to be a corruption of Bisila and Biaduas, the
Vistula and the Oder. (See his Beovulf, p. 100.)
The derivation of Mauringa, Maurungani must remain xinsettled. It
may mean the swampy moor land, from a root connected with O.H.G. mios,
English moss and mire (quite distinct from English moor1}. Other sug
gestions have been made, but they are mere guesses. Thus it is proposed
to derive the name from the root maur, an ant, on the following grounds :
Mauringa was in historic times left desert by its inhabitants : they must
have left the land because they were uncomfortable, and presumably over
crowded, the land was therefore called derisively " the ant heap"2.
Heinzel, after weighing this etymology and finding it wanting3, proposes
one, if possible, even more far-fetched. The Burgundians, according to a
Roman historian, knew themselves to have been of old of Roman stock4 :
hence perhaps they were of darker hue than the West Germanic folk5.
Now the Burgundians lived at one time east of the Elbe ; hence the land
east of the Elbe may have been inhabited by darker people, and hence have
derived its name from the Latin Maurus, a Moor, assuming this word to
have been early known in Germany — for which evidence is lacking.
(C) EASTGOTA.
Against the historical existence of Ostrogotha may be urged :
(1) His name, which looks mythical and eponymous, like King Dan of
the Danes, Hwicca of the Hwicce.
(2) The lack of satisfactory contemporary evidence.
But (1) On the other hand, we should expect Ostrogotha, if eponymous,
to come at the head of the pedigree. Only his having been an historical
character can account for his name coming so far down in the list. We
cannot pronounce a German chief mythical because his name seems
strangely appropriate to his career. It was intended that it should be
so, when he was named. The names Airmanareiks, Alareiks, Thiudareiks,
Gaisareiks exactly fit the conquerors who bore them. An heir to the
throne, born perhaps when national feeling was running high, might
well have been named The Ostrogoth. Theodoric named one of his
1 Zeuss (p. 472), Bluhme (p. 23), and Miillenhoff (Nordalb. Stud, i 140-41)
suggested this meaning : but they wished to connect with moor (O.H.G. muor),
which is linguistically impossible. See Miillenhoff, D.A. u, 97 ; Beovulf, 102 ;
Bruckner, 107 ; and cf. Ettmiiller d) 11.
2 This etymology was hinted at by Miillenhoff, D.A. 97, and has been
seriously adopted by Kossinna.
8 Ostgotische Heldewage, 24.
4 Ammianus, xxvin, 5, 11.
5 A most perilous assumption : for even if the Burgundians did boast of
a Roman origin, there is no reason for thinking that there was more ethno
graphical truth in this than in the Trojan origin of the British.
Appendix 237
daughters Ostrogotho1. An historic Gepid prince was named Ostrogothus2,
which is more difficult to account for.
(2) Whilst Jordanes and Cassiodorus are late in date, they are
remarkably consistent in their notice of Ostrogotha. Reckoning back
the generations of the Amal pedigree, we should expect to find him
where we actually do meet him, about the middle of the third century.
There seems then every reason to accept Ostrogotha as an historical
figure; and this has been done without demur by Gibbon3, Dahn4,
Miillenhoff6 and Hodgkin6.
Schiitte, Oldsagn, 152-3, should however be compared.
(D) THE JUTES.
Ttum \weold] Oefwulf,
Ytum = Early West Saxon letum, corresponding to which the Anglian
or Kentish form would be Eotum, lotum. These Yte, Iotas, Eote, Ytan
(corresponding to a Prim. Germ. Eutiz, Eutjoz, Eutjaniz) are identical
with the lutae, luti whom Bede mentions as conquering and colonizing
Kent, the Isle of Wight, and the adjacent parts of Hampshire. This is
shown by the fact that :
(1) The words Ytena (Corpus, Camb., MS.), Eota (Bodleian, Cotton,
Corpus Oxf., and Camb. Univ. MSB.) are used to render lutarum in the
Old English version of Bede's History, iv, 16, ea (gens) quce usque hodie in
prouinda Octidentalium Saxonum lutarum natio nominatur, posita contra
ipsam insulam Vectam.
(2) The neighbourhood of the New Forest was known as Ytene [i.e.
Ytena land] till at least the twelfth century. Florence of Worcester
speaks of William Rufus as having been slain in Nova Foresta quce
lingua Anglorum Ytene nuncupatur, and again in prouinda Jutarum in
Nova Foresta (Florentii Wigorn. Chron., ed. Thorpe, II, 45 ; I, 276).
Traces of the name are perhaps to be found in Ytingstoc (Bishop's
Stoke near Southampton — Birch, Cart. Sax. in, 1054 ; cf. Binz in
P.B.B., xx, 185). It is clear, then, that the Hampshire Jutae were
known as Yte (or perhaps Ytas, Ytan) : and we have Bede's evidence
that these were the same race as the Kentish folk (i, 15).
(3) lotum, lutna is used to translate lutis, lutarum in the passage
inserted in the A.-S. Chronicle (derived from Bede, i, 15).
On the other hand, in the O.E. Bede, lutarum in I, 15 is translated by
1 Jordanes, ed. Mommsen, cap. Lvni.
2 QiHTTplyorOos. Procopius, Bell. Gott. iv, 27.
3 i, 245. Gibbon does not mention Ostrogotha by name, but accepts as
historical Jordanes' account of the campaign.
4 Konige der Germanen, u, 1861, p. 54 ; Allgemeine Deutsche Biographie,
XXIV.
5 Z.f.d.A. ix, 136. 6 Italy, i, 48.
Widsiih
Geata, which has led to the theory that the Jutlanders and the Jutish
conquerors of England are to be identified with the Geatas of Beowulf.
Fahlbeck, who holds this view, supposes the name Tte to be derived from
Vecta, and to have been merely the name adopted by the Jutish settlers
in the Isle of Wight and on the adjoining coast : it was only later, he
thinks, that the word Yte came to be regarded as synonymous with all
Jutae, and thus became applicable also to the Kentishmen (see Fahlbeck,
Beovulfsqvadet, 46-48). This theory is hardly tenable ; we should have
to locate Gefwulf and his Yte in England, and attribute to the catalogue
of kings in Widsith a date and character quite different from that to which
all the other evidence points. For the kings in this catalogue all belong
to the continental hero cycles of the fourth to the sixth centuries, not to
the insular English cycles of the seventh. The Yte of Widsith are almost
certainly a continental people, as is shown by their being grouped with
Wsernas, Eowan and Frisians.
Fahlbeck's view is unsatisfactory, for it is based upon one passage in
the O.E. Bede, where Jutarum is rendered Geata, whilst it is refuted by
the other passage in the O.E. Bede, by the Chronicle, and by Florence,
where Jutarum is translated by Ytena, or its Anglian or Kentish equiva
lent, Eota, lotna. Of. too Schiick, 16-18.
We may accept, then, as fairly certain, the identification of the Yte,
Eotas, with the lutae of Bede. They are probably also identical with the
Eutii, Euthiones who are mentioned as having made some kind of sub
mission to the Frankish kings 1 in the sixth century. That these Frankish
references are to the Yte dwelling on the English side of the channel,
rather than to those who remained in their original continental home, has
been suggested by Weiland (p. 36) and Bremer (Zachers Z.f.d.Ph. xxv,
130), and is quite likely.
Where this original continental home was situated is not so clear.
Bede places the old Anglian country inter provincial lutarum et Saxonum
(i, 15), he must therefore have regarded the lutae as dwelling north of
Angel, i.e., in the modern Jutland. It has been urged by Jessen
(Undersfigelser til Nordisk Oldhistorie, K0benhavn, 1862, 55), Moller
(V.E. 88; A.f.d.A. xxn, 159), Weiland (36), Mullenhoff (Beovulf, 98),
and more recently by Siebs (Pauls Grdr. (2) I, 1158) and Heuser
(Review of Biilbring in the Anzeiger to the Indogermanische Forschungen,
xiv, 1903, 26-30), that the character of the Old Kentish dialect makes
Bede's statement incredible, and that Old Kentish dialectal peculiarities
1 '...subactis Thoringiis . . . cum Saxonibus, Euciis, qui se nobis voluntate
propria tradiderunt.' Letter of Theudebert to Justinian, Epistolae Merowingici
Aevi, i, 133 in M.G.H. (between A.D. 534 and 547).
Quern Geta, Vasco tremunt, Danus, Euthio, Saxo, Britannus
Cum patre quos acie te domitasse patet.
Venantius Fortunatus in praise of Chilperic, about the year 580. See Car-
minum nc, 1, 73, ed. Leo, in M.G.H. Auct. Ant. iv, 1881.
Appendix 239
would lead us to look for the original home of Hengest's warriors rather
in Friesland than in Jutland1. This assertion as to the close connection
between Kentish and Frisian has grown to be a dogma. (See however to
the contrary Chadwick (67) and Bjorkman in Engl. Stud, xxxix, 1908,
361.) The matter needs investigation, but the coincidences between
Kentish and Frisian have certainly been exaggerated, and seem insufficient
without further corroboration to invalidate Bede's statement.
It has further been objected that any connection between fte, Eotas,
lutae, on the one hand, and Old Norse Jotar (Jutlanders, the luti of
Saxo), is phonetically impossible (Moller V.E., 88). But this objection
arises from the assumption that the original name began with a j. This
should, of course, be dropped in Scandinavian and represented by a g in
Old English (cf. Goth, juggs, Icel. ungr, O.E. geong). It was however
pointed out by P. A. Munch as long ago as 1848 that the Old Norse
Jotar, as well as the O.E. forms, can both be traced back to a primitive
form beginning not with Ju but with Eu, and that the evidence of
Latin writers is that this was the original form.
Admitting this, Moller urges, as a last objection (private communication
quoted by Kossinna, I.F. vn, 293) that at any rate the modern Danish
form Jyder, if a pure Danish word, must go back to a form in J such as
*Jeutiones : it could only come from a form *Eutiones if it were a North
Frisian loan word. Undue weight seems to have been attached to this
objection : for ex hypothesi the mod. Danish Jyder would not be a pure
Danish word, and the objection therefore falls to the ground. For it is
not argued that Bede's lutae were racially identical with the modern
Jutlanders, who are of Scandinavian stock: but only that the latter
adopted their name from the earlier "Anglo-Frisian" inhabitants of
Jutland, whose country, after the emigration of Hengest to England,
they settled, and the remnants of whom they assimilated. On the Danish
development of the word, see Jordan, 139.
Assuming, then, that the original name of the tribe was Eutiz (or
Eutjds, Eutjaniz)2, we should get in the West Germanic dialects through
the influence of the front vowel in the second syllable, lutiz, etc. (cf.
Wright, Primer of Gothic, § 61). In Bede's time the lu would still be
preserved in Northern English (cf. Sievers, Zwr Geschickte der ags.
Diphthonge, in P.B.B. xvni, 411-416), and he would write, quite
correctly, lutae. lutiz would give Anglian 'Eote, West Saxon lete, Yte;
whilst, on Mailer's own showing, the Scandinavian tribes, borrowing the
1 The holders of this view admit the possibility of the Hampshire men,
of whose original dialect we know nothing, having come from Jutland ; they
only deny that the Kentish men can have done so (Moller, V.E. 88, footnote :
Miillenhoff, Beovulf 98). Jessen argues that the settlers of the Isle of Wight
were of Scandinavian stock.
a Ten Brink (204) and Stevenson (167) are probably right in supposing the
word an i-stem. Much (P.B.B. xvn, 208) suggests a jo-stem, Munch an
n-stem. Euciis, Tie, point to an i-stem : Euthio, Ytena to an n-stem. Perhaps
both existed side by side, cf. Seaxan and Seaxe, Miercna and Mierce.
Widsith
name from West Germanic sources, would naturally alter it to O.N. Jotar,
Danish Jyder.
To sum up. (1) There is no sufficient ground for doubt that the Yte
are identical with Bede's luti, lutae, the conquerors of Kent, and with
the Eutii, Euthiones indicated by continental authorities.
(2) Holler's attempt to prove that the original tribal name underlying
all these later forms cannot be identical with that from which the name of
Jutland is derived has not succeeded. As to the original home of these
Yte, we are then left on the one hand with Bede's statement that they
dwelt north of the Angles, and on the other with the contention that
the close similarity between the Old Kentish dialect and the Frisian
speech would lead us to put the ancestors of the Kentish men in the
closest proximity to Friesland. But in the fifth century dialects hardly
distinguishable may well have been spoken on all the North Sea coast,
from Friesland to Jutland. Our knowledge of the mutual relations of
these tribes and of their dialects is too imperfect for us to argue with
certainty.
The subject is one on which agreement will, in all probability, never
be reached. The English historian, impressed with Bede's reliability, will
probably continue to repeat his statement, without any doubt or hesitation
(cf. e.g. Sir James Eamsay's Foundations of England, 1898, 1, 120). Many
philologists, on the other hand, will probably continue to believe that
linguistic evidence makes Bede's statement incredible. Other philologists,
however, have accepted it ; e.g. Zeuss (501), Grimm (Gf.D.S., 738), Ten
Brink (Beowulf, 204), Much (in P.B.B. xvn, 209), Bremer (in Pauls
Grdr. (2) in, 856), Biilbring (Altenglisches Elementarbuch, 1902, 5).
Stevenson, in his edition of Asser, 1904 (pp. 166-70), seems doubtful,
and doubt has been more strongly expressed by Sarrazin (Beowulf-
Studien, 28) and Erdmann (Heimat der Angeln, 40) l. Among those
who have argued against Bede's location of the lutae in Jutland,
mention should be made, in addition to the references to Jessen,
Moller, Mullenhoff, Weiland, Siebs, and Heuser, given above, to
Kossinna's Ethnologische Stellung der Ostgermanen in I.F., vu, 276, etc.
The question cannot be settled, because it depends upon the relative
weight of two quantities difficult to determine, (1) Bede's reliability in
dealing with events which occurred more than two centuries before his
birth, (2) the extent to which we are justified in arguing, granting the
disputed similarity of Kentish and Frisian in historic times, that four
centuries before those times the ancestors of these peoples must necessarily
have been immediate neighbours. To the present writer it seems that,
whilst the evidence upon which Bede based his statement that the lutae
1 Cf. also Erdmann, Om folknamnen Gotar och Goter, in the A,T,f.S. xi, 4,
19-24.
Appendix 241
dwelt north of the Angles may have been insufficient, the evidence by
which it is sought to refute this statement indubitably is insufficient, and
that Bede's statement accordingly holds the field.
Those who would make the Jutes originally neighbours of the Frisians
are not agreed whether to put them on the north-eastern border (Siebs in
Pauls Grdr. (2) i, 1157) or on the south-west, in the Netherlands (Holier,
Heuser). Hoops' suggestion that the Jutes may have come from Jutland
in the first instance, but may have settled temporarily in the Netherlands
( Waldbaume, 585), is an ingenious compromise, which derives considerable
support from the fact that there is some evidence that the Angles and
Saxons attacked Britain from such temporary settlements in the Nether-
lauds1. (See Note E : The original homes of the Angli.)
(E) THE ORIGINAL HOMES OF THE ANGLI AND VARINI
(ENGLE AND \\OERNAS).
The current view, that the original home of the Angles is to be sought
in and around the modern Schleswig, rests upon the assertion of Bede2
that the Angles came from Angel, and the apparently independent, because
slightly dissimilar, statement of Alfred. The A.S. Chronicle, Ethelwerd,
and William of Malmesbury, depend upon Bede ; but the reference to the
Insula Oghgul in the Hisioria Brittonum is probably independent and
corroborative. As has been pointed out above, it is clear that the
Widsith poet regards the Angles as a people dwelling in the neighbour
hood of Schleswig. For the poem makes Offa king of Ongel defend his
frontier at Fifeldor — the river Eider. If it can be proved that the Angles
came, not from the neighbourhood of Schleswig, but from the interior of
Germany, then we must abandon any belief in Widsith as in the main
early or authoritative, although we might still hold that it incorporated
early fragments.
It was Zeuss who first dismissed the connection of the Angles with
Angel as a piece of false etymology, resting upon the gossip of sailors.
Zeuss placed the home of the Angles on the lower Saale3. To this con
clusion he was led, partly by the many undoubted traces of the existence
of Anglian or kindred tribes in the Saale neighbourhood4, and partly by
1 In a discussion of the Jutes it is better to leave out of the question
(1) the Eudoses of Tacitus. These may be identical with the Eutii, lutae,
Yte (as Miillenhoff supposed, Nordalb. Stud, i, 112-119, though he afterwards
abandoned the view). But if they are, the d, Eudoses, not Eut-, remains in
explicable.
(2) the *Eotenas mentioned in the Finnesburh lay in Beowulf. Holder,
in his Beowulf (128), and others have wished to connect these with the Eotas-
lutae. But this needs investigation.
2 See above, p. 74. 8 p. 496.
4 Since strengthened : cf. Bremer in P.B.B. ix, 579.
o. 16
242 Widsith
the desire which he often betrays of making his geography square with
that of Ptolemy1. In view of the authority of Zeuss, whose work on early
German geography is still what Kemble declared it to be — indispensable —
and of the fact that the most elaborate investigation of the origin of the
Angles, that of Axel Erdmann2, endorses his view, it ought not to be
assumed without investigation that the location of King Offa given in
Widsith is necessarily correct.
As to the arguments which led Zeuss to place the Angles in the centre
of Germany : it is certain that there was a settlement, on the Saale, in
the neighbourhood of the modern Merseburg and Leipzic, of a people at
any rate closely akin to the Angles, speaking a dialect similar to that of
the Angles and Frisians. The Merseburg Glosses3 written down at the
beginning of the eleventh century, and still preserved in the Cathedral
library at Merseburg, show traces of the " Anglo-Frisian " peculiarities.
So do the forms of the German proper names adopted by the local
chronicler, Dietmar von Walbeck, whose history, compiled about the
same time that the glosses were written, and extant in a MS. partly
written, partly revised, by himself, is valuable evidence as to the
dialect of the district in the 10th century. Further, the old names of
the districts, JEngelin, Werinofeld, near Merseburg, point, not merely to
a general " Anglo- Frisian " element in the population, but to the settle
ment here, on the borders of Thuringia, of bodies of Angles and Varini.
In two places in the neighbourhood of Leipzic the name Wernsdorf survives
to this day*.
But whilst admitting this settlement of Anglo-Frisian people in the
inland districts, it would require the very heaviest evidence before we
could agree with Zeuss and Erdmann in making this the original home of
the Angles, from which they conquered England. Such a theory is, we
have seen, opposed to the express evidence of Bede, Alfred, and the
Historia Brittonum. To make the Angles an inland people is also
opposed to the evidence of Tacitus, who makes them share in the
worship of Nerthus on an isle of the ocean; as well as to the evidence
of Beowulf, which places Offa I betwixt the seas, be seem tweonum. Zeuss'
theory is indeed, if we may say so without disrespect to the memory of so
great a scholar, contrary to common sense. To give a radius of over two
hundred miles to the area of marsh and forest which had to be covered by
the car of the goddess Nerthus is surely grossly to over-estimate the powers
1 Cf. Miillenhoff's criticism of Zeuss in the Nordalb. Stud, i, 112.
2 Ueber die Heimat und den Namen der Angeln, Upsala, 1890. So too
Steenstrup in Danmarks Riges Historie, i, 76-77.
8 Edited by H. Leyser in the Z.f.d.A. in, 280; and again by Bezzenberger,
Z.f.d.Ph. vi, 291. Heyne (Kleiners Altniederdeutsche Denkmaler) first noticed
the close dialectal resemblance of Dietmar's proper names to the glosses ; and
Bremer (P.B.B. ix, 579) first identified the dialect as "Anglo-Frisian." See also
Seelmann, Thietmar von Merseburg, die Merseburger Glossen und das Merse-
burger Totenbuch, in J.d.V.f.n.S. xn, 89.
4 Schroder, Lehrbuch der deutschen Rechtsgeschichte, 3to Auf. 1898, p. 244.
Appendix 243
of the primitive Germanic cow. Besides, it seems incredible that the men
who sailed across the North Sea to conquer England should have come,
not from the opposite coast, but from a district more than two hundred
miles inland. Erdmann seeks to surmount this last difficulty by sup
posing that the Angles may have been accustomed to use the Elbe for
communication (pp. 23-31). But can example be found anywhere of an
inland folk, even if settled upon a great river, having chosen the calling of
sea-robbers? Erdmann1 quotes the Heruli with their ravaging of the
Aegean and Euxine coast. But this is not parallel, for whatever be the
origin of the Heruli, they were probably in the first place a sea folk5'. The
nearest instance would be that of Gaiseric and his Vandals. They were
certainly landsmen, yet, after their conquest of Carthage, they made them
selves dreaded throughout the whole Mediterranean. But, in Carthage,
Gaiseric had won one of the four great ports of the ancient world, with all
its ships, docks, galley slaves and pilots3. The case is therefore not
parallel ; for the men who conquered England had to rely upon their
own seamanship ; and they needed more seamanship than was required to
row a dug-out down the Elbe, before they could conduct a campaign across
the North Sea
Again, the character of the Anglian speech, with its affinities to Frisian
and to Scandinavian, would be unintelligible on the hypothesis of Zeuss.
(Cf. Jordan, Eigentumlichkeiten des anglischen Wortschatzes, 1906, 114;
Jordan in Verhandlungen, 139-140 ; Mb'ller in A.f.d.A. xxn, 147, etc.}
Yet the evidence for placing the original home of the Angles on the
Saale, near Merseburg, so far from being of the overwhelming kind necessary
to overcome difficulties like these, is of the very slenderest. It rests solely
upon the words of Ptolemy :
" The greatest of the tribes dwelling in the interior are the tribe of
the Sueboi Aggeiloi, who lie east of the Laggobardoi, stretching towards
the north as far as the middle of the river Elbe ; and the tribe of the
Sueboi Semnones, who extend from that portion of the river Elbe
eastwards4."
By the Suevi Angili Ptolemy pretty clearly means the Angles (for the
Angles, according to Tacitus, formed with the Langobardi and the Semuones
part of the great Suevic confederation). The Suevi Langobardi have already
been located by Ptolemy below the Sygambri on the Rhine : whilst the seats
of the Semnones we know from abundant evidence. Ptolemy's Sueboi
Aggeiloi, between these two, would still be fifty miles distant from the
1 p. 32.
2 Dani Herulos propriis sedibus expulerunt, Jordanes in. See note to Eolum
(1. 87).
3 See Gibbon, ed. Bury, in, 410.
4 TWV 5£ frrbs Kal /j.effayeluv tQv&v niyiara i^v t<m. T(> re ruv "Zirtifiwv r&v
A.yyei\wv, 01 dffiv dvaroXiKurepoi r&v Aayyofidpduv dvarelvovres Trpos ray Apitrovs
fJ^XP1 r<*>v tdffuv TOV "AX/3ios irorafj-ov, Kal ri> TLOV Suij/Saw rCiv 2e/JU'6t>i*v, oirices
fifTa rbv "AXfiiv dwo TOV elp-q/dvov ^pous ir/ods dvaToXdj ^XP1 T°*>
.... Geographia n, 11, 8.
16—2
244 Widsith
district of Merseburg and Leipzig. This essential point is overlooked by
Erdmann, who speaks as if the account of Ptolemy (and, indeed, of Tacitus)
placed the Angles on the Elbe and Saale (p. 24). Yet nearer than fifty miles
no amount of stretching of Ptolemy's words will bring them ; for the
Merseburg district is south, not west, of the ancient home of the Semnones.
But, further, Ptolemy is here so confused and uncertain that he has
puzzled generations of scholars. The explanation given by a recent editor,
Muller, seems convincing1. Ptolemy, it must be remembered, accurate and
well informed on the whole as he was, wrote in Egypt, compiling his notes
of German geography from various more or less imperfect statements. He
knew, and reported quite correctly, that the Angles dwelt N.E. of the
Lombards : but as to the Lombards he failed to reconcile his authorities.
Under the name of the Aa/t»co£ap8oi he placed them quite rightly "next to
the Angrivarii," i.e. on the left bank of the lower Elbe. But meeting them,
in some other record, described as the Suevi Langobardi — he did not
recognize them as the same people. He could not put these Suevi Lango
bardi in their right place, which he was reserving for their shadows, the
Laccobardi. Consequently they are misplaced, on the right bank of the
Rhine ; and this, in turn, leads inevitably to a dislocation of the Suevi
Angili, whose position is denned with reference to the Suevi Langobardi.
Yet just because in the second century a puzzled Egyptian geographer
failed to unravel the truth from conflicting accounts, we must not give the
lie to Bede and Alfred. When we adopt Miiller's explanation, and perceive
that the Laccobardi and the Suevi Langobardi are one and the same,
the Angles drop at once into the right place, N.E. of the Lombards, and
consequently on the sea coast, in conformity with the evidence of Tacitus,
Bede, Alfred, Beowulf, and the Historia Brittonum.
But though Ptolemy's evidence for an original Anglian home in central
Germany falls, on examination, to the ground, the linguistic evidence, as we
have seen, and the evidence of place names proves that there was some kind
of a settlement of Anglian and kindred tribes on the banks of the Saale,
on the northern borders of the ancient Thuringia. How this settlement
came about we do not know2, but we seem to catch a glimpse of these
1 See Claudii Ptolemaei Geographia, ed. C. Miillerus i, 1, 258. Paris,
Didot, 1883. " E boreali ilia regione Anglos in mediam Germaniam penetrasse,
aut priscas eorum sedes ibi ad mediam Albim fuisse, e Ptolemaeo multi colle-
gerunt, quern ego hoc quoque loco falsa exhibere censeo. Becte quidem
Ptolemseus Angilos a Langobardis ortum et boream versus collocaverit, sed
quemadmodum Langobardi isti ponendi erant eo in loco ubi Laccobardos
posuit, sic etiam Angli a Laccobardis s. Langobardis ortum boreamque versus
inter hos et Saxones, collocandi fuerint. " Cf. also Mullenhoff on Ptolemy in
Nordalb. Stud, i, 112 ; and Z.f.d.A. ix, 233.
2 The obvious conjecture is that of an immigration of Engle and Wsernas
from the districts nearer tbe mouth of the Elbe, in perhaps the fifth century —
hardly earlier, as the immigration must have taken place after the Anglo-
Frisian idiosyncracies had become marked. See Bremer in P.B.B. ix, 579.
Seelmann (p. 90) denies the immigration so far as the Angles are concerned,
Appendix 245
"Thuringian" Angles and Varini once about the year 500, and again
about the year 800.
(1) Amongst Cassiodorus' Varice1 there is extant a letter from Theodo-
ric, the Ostrogoth, to the kings of the Heruli, Warni, and Thuringi,
suggesting an alliance. Clovis, king of the Franks, was at this time
pressing hard upon Theodoric's friends and allies, the Visigoths, and it was
Theodoric's aim to bring about a coalition which should prevent him
pursuing his aggressive schemes.
(2) There is also a code of laws, dating probably from the time of
Charles the Great, three centuries after Theodoric's letter, which bears the
superscription Lex Angliorum et Werinorum hoc est Thuringorum. This
" Law of the Thuringian Angles and Varini " is said by students of com
parative Germanic law to show traces of kinship with English, and
particularly with Kentish law ; but in the main it has taken the form
impressed upon it by the Frankish influence under which it has passed2.
(3) Further confirmation of the settlement of the Varini (Waernas) in
ancient North Thuringia has been found in the frequent occurrence of
place names ending in leben within that district. This form is not found
elsewhere in Germany till we come to Schleswig, where it occurs in the form
lev. This fact, when first pointed out3, was taken to support the theory of
an Anglian settlement in Thuringia. But it has been shown more recently
that it is only in Northern Schleswig that the names in lev are found, and
that they extend over all modern Denmark, and, in the form lof, over parts
of the southern coast of Sweden4. In Angeln and the islands immediately
adjoining these names are not found : and they are found only rarely
in England. These place names then point, in all probability, to settle
ments, not of the Angles proper but of their northern neighbours, Varini
and Jutes. The occasional occurrence of English place names in lew, ICKW
is to be attributed to settlements of Jutes and Wsernas in England.
We, may, then, take it as certain that there was, from at least the early
sixth to the early ninth century, upon the northern borders of Thuringia,
a semi-independent kingdom of the Wsernas, and, adjoining it, settlements
of the Engle and Heruli.
Unfortunately, the question of the home of these " Thuringian " Angli
and Varini has been complicated by the fact that there were two Thuringias :
the great kingdom in central Germany, and a smaller Thoringia near the
and regards the Saale- Angles as ancient inhabitants of the country, not immi
grants from Schleswig. But this is contrary to probability, and rests merely
upon Ptolemy's statement as to Angles in the neighbourhood (and certainly not
the immediate neighbourhood) in the second century.
i m, 3. 2 Chadwick, 81, 108.
3 By P. Cassel, Ueber Thiiringische Ortsnamen, Erfurt, 1856 (Berichte der
Erfurter Akademie).
4 See Seelmann, Die Ortsnamenendung "leben," J.d. V.f.n.S. xii, 7-27. It
is conceivable, though hardly likely, that this type of place name cropped up
independently in the different districts, as Erdmann suggests (p. 57).
246 Widsith
mouth of the Rhine, in the neighbourhood of the modern Dordrecht 1. Hence
arose the theory that the Anglii and Werini referred to in the Lex, and
the Guarni and Thoringi addressed by Theodoric, dwelt in the Netherlands2.
A further development of this theory would make these Netherland Angles
the source of the English name. According to this view, settlers from the
Anglian nation at the mouth of the Rhine may have taken part in the
invasion of England in sufficient quantities to have given their name to
parts of the conquered country, and it may be to them rather than to
any connection with the Angel in Schleswig that the name of England
is due3.
Now here we must discriminate carefully. That there were in the
Netherlands settlements, small or great, of Engle and Wsernas, of Saxons
and Jutes, is possible enough ; and from the shores of the Low Country
many of the attacking fleets may have set out. But this does not help us
with regard to the original home of the Engle and Waernas, for the state
ments of Ptolemy and Tacitus make it quite clear that Angli and Varini
were not settled in the Netherlands in the first and second centuries. These
Angles of the Low Countries must then have been immigrants from an
older home, wherever that home may have been. But that the Warni
to whom Theodoric wrote, and the Anglii and Werini of the Lex,
may have dwelt in the Netherlands, and that these may have been
intimately connected with the settlers in England, is possible enough.
Direct evidence, it is true, is wanting. For the Angles there is none,
beyond a corrupt passage in Adam of Bremen4, who in any case is an
authority too late to count for much in the period we are considering.
1 See note to pyringum (11. 30, 86).
2 This theory was first advanced by H. Miiller (Der Lex Salica und der Lex
Angliorum et Werinorum Alter und Heimat, 1840, pp. 120 and 121). Miiller
was followed by P. C. Molhuysen, De Anglen in Nederland, in the Bijdragen
voor Vaderlandsche Geschiedenis en Oudheidkunde, in, pp. 50-72 (1842). Waitz
tentatively approved this conjecture (Das alte Recht der Salischen Franken,
1846, p. 50).
Arguments against the theory of an Anglian settlement in the Netherlands
will be found in a review of Muller's work by the Freiherr v. Eichthofen in the
Kritische Jahrbiicher filr deutsche Bechtswissenschaft, 1841, pp. 996-1018. See
especially p. 1015, with its very reasonable warning against drawing any con
clusion from the occurrence of small isolated places compounded with the name
Engel. " In a district where the bulk of the population were Angles, a single
spot would hardly be called by their name." Mr Inman has plausibly sug
gested that "the occurrence of Aengil — Engel — in relation to places in Belgium
and Holland (Felwe, Ghent, Texandria, Texel, Tournay, Utrecht) is possibly in
connection with Anglian missions of the 7th or 8th century."
3 "Ved Binmundingerne bode Angler med Varner, altsa hasrMger udgaede
fra Anglerne og Varnerne inde i Tyskland. Hovdinger herfra i ledtog med de
egenlige Sakser (Friser) kunde flytte Angle-navnet til England." Undersflgelser
til Nordisk Oldhistorie af C. A. E. Jessen, Ktfbenhavn, 1862, p. 51. But Jessen
thinks it more likely that the name sprang up independently in England.
4 Igitur Saxones primo circa Eenum sedes habebant [et vocati sunt Angli],
Adam of Bremen,, i, 3, ed. Lappenberg in Pertz (fol.), SS. vn, p. 285. The
words are wanting in the best MS. and were accordingly bracketed by the editor.
Indeed "they can hardly be taken seriously" (Chadwick, p. 111).
Appendix 247
For the Wsernas we have the statement of Procopius that the Rhine
divides the Varini from the Franks1; but Procopius' ideas as to the
geography of these regions are confused in the extreme. The strongest
evidence is of an indirect nature. Thus, it has been urged2 that the Old
English loan words from the Latin are, on the one hand, of a kind which
could not have been borrowed in the remote continental home in Schleswig,
but that, on the other hand, they show such close relationship to con
tinental loan words as to preclude the possibility of their having been
borrowed subsequent to the settlement in Britain, and that consequently
the Angles must have been settled for some time in a continental home
subject to Latin influence. So that it is quite possible that those scholars
are right who believe that there were large settlements of Engle and
Waernas in the Netherlands3: it is certain that there were also such
settlements far inland, on the upper Elbe and Saale. But that neither of
these could have been the original home of the Angles is clear from the
evidence of Tacitus alone, had we nothing more. " We may boldly main
tain that in these dark ages no similar fact is better attested than the origin
of the English Angles from Angel in Schleswig4." In these words the
controversy was summed up some twenty years ago ; and though in every
generation some good scholar revives the heresy of Zeuss, and brings his
Angles from the inland, none have yet succeeded in finding any satisfactory
justification for this eccentric view.
Here then, as in every other point of old Germanic tradition where we
are able to control him, we find the poet of Widsith to be well informed.
1 E.G. iv, 20. See note to Wernum (I. 25).
2 Hoops' Waldbaume u. Kulturpflanzen (esp. Kap. xiv, Die kontincntale
Heimat der Angelsachsen und die romische Kultur, 566-589).
3 Mullenhoff in Nordalb. Stud, i, 130-133, 1844, accepted Miiller's view
of Anglian settlements in the Netherlands : so did Moller (V.E. 18). Holler's
arguments in favour of England having been invaded from the Netherlands will
be found in A.f.d.A. xxn, 152. Bremer, in Pauls Grdr. (3) n, 854, also places
the Lex Angliorum in the Netherlands, and supposes the advance guard of the
Angles to have had their base there, though the bulk of their later followers
were drawn from Schleswig. On the other hand, that the Lex belongs to the
Angles of the Elbe and Saale is urged by many scholars. Schroeder (Lehrbuch
der deutschen Rechtsgeschichte, 1902, p. 248) regards this as the only tenable
view. Though the Lex must belong to one or the other locality, there is nothing
inconsistent in supposing that there were settlements of Engle and Wsernas both
in the Thoringia of the Netherlands and in the main Thuringia (so Erdmann,
Heimat der Angeln, p. 28). That the evidence of loan words points to Saxons
and Jutes having invaded England from the lower Rhine has also been argued
by Heuser, I.F. xiv, Anzeiger, p. 30 (1903).
4 Weiland, Die Angeln, 1889, p. 18.
248
(F) ISTE.
The Iste are apparently identical with the Aestii of Tacitus. What
were the linguistic and racial affinities of these Aestii is not certain, but
their situation is fixed. They dwelt on the "Amber Coast" of the Baltic,
which extends from the Frische Haff in East Prussia in a north-easterly
direction over the frontier into the "Baltic Provinces" of Russia. Tacitus
describes the Aestii as being, in their religious rites and dress, like the
Suevi, and though he notes that their tongue resembles the British, he
classes them among the Germans. They gather amber on the coast
sed et mare scrutantur, ac soli omnium sudnum quod ipsi glesum vacant
inter vada atque in ipso littore legunt....Ipsis in nullo usu, rude legitur,
informe perfertur, pretiumque mir -antes accipiunt1.
In a letter of Cassiodorus (Hestis Theodericus rex2) the Hesti are
thanked by the Gothic king for a gift of amber, and edified by his minister
with much information as to its nature, which Cassiodorus has taken out
of the Germania. The Aesti are also mentioned by Jordanes as occupying
the sea coast a little to the east of the Vistula : he characterizes them as
pacatum hominum genus, and mentions them as having formed part of
Ermanaric's empire3.
The correct form of the word in Latin writers is uncertain : the MSS.
of Tacitus fluctuate : apparently we should read Aestii, the Ae representing,
as often, a Germanic Ai (Kluge in Pauls Ordr. (2) I, 357). Einhard calls
them AM, Haisti ( Vita Caroli in Pertz (fol.), SS. n, 449). The Gothic
form of the name would be probably *Aistjus, pointing to a Prim. Germ.
Aisteues* ; or perhaps Gothic *Aisteis6. With these Gothic forms the Icel.
Eistir6 would agree : the corresponding English form would be *^Este,
which however is nowhere found.
Two forms of the word are found in the O.E. Orosius. (1) Osti, a form
no doubt borrowed from the Germans of the Continent and due to false
analogy with dst, "east"7. (2) Este, the form used in the account of
Wulf stan's voyage. It has even been supposed by some scholars8 that
1 Germania, XLV. 2 Variae, v, 2.
3 Aestorum quoque similiter nationem, qui longissimam ripam Oceani Ger-
manici insident, idem ipse prudentia et uirtute subegit (ed. Mommsen, xxm).
4 Cf. Erdmann, p. 90 ; Miillenhoff, D.A. n, 13.
B Grimm, G.D.S. 719 (fourth ed. 500).
8 But note that the Eistir of the sagas do not occupy the same country as
the Aestii of Tacitus. These Eistir live far to the north of the "Amber Coast,"
in the modern province of Esthonia. That the O.N. Eistland is limited to
Esthonia and did not include the neighbouring Courland or Livonia is clear.
Cf. Hjd Garftariki liggja lond }>essi.... Eistland, Lifland, K&rland. (Annota
tions geographiques tiroes du livre de Hauk Erlendsen in Antiquites russes, n,
438.)
7 Cf. Miillenhoff, D.A. u, 13. The Baltic is called in the Orosius the
Ost-Sce.
8 e.g. Latham in his edition of the Germania. 1851, 166, etc. ; Bielenstein,
300, 373.
Appendix 243
the word means " Easterlings," the Eastern folk ; but that there is any
original etymological connection with " east " is impossible, in view of the
Latin forms (Aist- never Aust-), and of the fact that the Icelandic sagas
generally give the form Eistland, only exceptionally Eystland (see Fornaldar
Sogur, 1829, i, 509). But that a popular etymology sprang up, connecting
Esth and East, is clear ; such an etymology would be almost inevitable in
view of such constantly recurring phrases as " Esthland in the East."
(Cf. Fornm. S. iv, 162 j vm, 272.)
The same false etymology accounts, as has been stated, for the O.E.
form Osti, and apparently for Este as well. It also accounts for our not
finding the phonetically correct form jEste (<Astiz), but instead the
form in our text, Iste (< leste < Eastiz).
With regard to the racial affinities of the Aestii : Grimm, following
Tacitus, supposed them to have been Germans, and claimed in support of
his theory that in Widsith the Iste are mentioned together with German
tribes. This argument, however, is based upon an interpretation of Eolas
and Idumingas which is doubtful in the first case, and almost certainly
wrong in the second. Erdmann also (p. 93) supposes the Aestii to have
been originally a Gothic stock, whence the embassy to Theodoric the
Goth1. Both Grimm and Erdmann admit that this German stock was
later swamped by Finnish and Lithuanian elements, which appropriated
to themselves the name of the original German tribe2. More generally,
however, it has been supposed that the Aestii were from the beginning
either Finnish or Lithuanian3. According to Bremer they were Finnish4,
but the greater number of scholars have agreed in regarding them as
members of that Lithuanian family which to this day occupies the
country where we first meet with the Aestii of Tacitus or Cassiodorus.
"On this Baltic coast," says Zeuss, "has dwelt through the centuries a
race which must be distinguished from both its neighbours (Germans and
Slavs) and which, through all the great movements which have taken
place around it, has spread but a little over its original borders. This
race is commonly called, from one of its branches, the Lithuanian. Its
true collective name is that of the Aestii5." Accordingly a certain group
1 But, if such a kinship had existed, is it not likely that some reference
would have been made to it in Theodoric's letter of thanks ?
2 It is clear that when Wulfstan visited the Este their language and customs
were not German.
3 As Tacitus can only have known the Aestii by repute, his statement that
their speech was British is of little positive value ; yet it serves to show that
they must have differed in speech from their German neighbours.
4 Pauls Grdr.(z) in, 753; cf. also Bremer's review of Mullenhoff's D.A. in
the Litteraturblatt f. germ. u. rom. Philol., 1888, 436 ; Kossinna, Zeitschrift fiir
Ethnologic, xxxiv, 214, etc. Much (Hoops Reallexikon, 54) argues from the
particulars given by Tacitus that the Aestii must have been Lithuanians, not
Finns.
8 See Zeuss, 268, and cf. Mullenhoff, D.A. n, 11-34.
250 Widsith
of writers have used the word Aistisch as the general term for this
Lithuanian family. Yet it is not certain that the Aestii were not
Finns : and it is certain that in modern times, through the Norse
Eistir, the word Esth, Esthonian, has come to signify a non-Lithuanian
folk, more nearly allied to the Finns. Probably the name has been
transferred from the Lithuanians to those tribes north of them who are
now called Esths. But we cannot be certain.
If the word was at one time supposed by the Scandinavians to signify
" Eastern men " we can the more easily understand how it lost its racial
signification, and came to be transferred by them from one eastern people
to another, from the Lithuanians of the Amber Coast to the Finnish
tribes of Esthonia.
The problem of the racial affinities of the Aestii is necessarily bound
up with the larger question whether the present Baltic Provinces of
Russia were, in the early centuries of our era, inhabited by a Livonian
(Finnish) or a Lithuanian or Lettish (Indo-Germanic) people. A summary
of this controversy, with full references, will be found in Bielenstein, Die
Orenzen des Lettischen Volksstammes, St Petersburg, 1892, 348-378.
The probability is, then, that O.E. form Iste is due to false analogy
with East ; that it refers to the Aestii, and that the Iste, Aestii were a
Lithuanian people. But we cannot be certain.
(G) IDUMINGAS.
J. Grimm (G.D.S., 500) suggested the emendation Eodingum, which he
interpreted as the luthungi, recorded by Idatius and other chroniclers.
(See Hydatii Continuatio in Momm sen's Chronica Minora [A.D. 430],
luthungi per eum [Aetiurn] debellantur ; and cf. Zeuss, 312-314.)
Grimm would further identify these Eodingas with the [NJuithones of
Tacitus. He abandoned this conjecture, however, preferring to identify
the Idumingas with the Ydumaei (" Idumingum rechtfertigt sich durch
die Ydumei bei Zeuss, 682," a MS. note of Grimm's, incorporated by
Miillenhoff in the posthumous edition of the G.D.S., 1867).
These Ydumaei are mentioned in the Chronicle of Henry of Ymera
(cf. Henrici Chronicon Lyvoniae ex rec. W. Arndt, Hannoverae, 1874;
Heinrich's von Lettland Livlandische Chronik, iibersetzt von E. Pabst,
especially the footnote to p. 77). The identification of the Idumingas
with the Ydumaei had been made by Ettmuller in 1839 (see pp. 24-5),
and later, perhaps independently, by F. J. Wiedemann (see his Intro
duction to J. A. Sj^gren's Livische Grammatik, St Petersburg, 1861,
p. xxix), who was followed by Koskinnen (Sur I'antiquitd des Lives en
Livonie, in the Acta Societatis Sdentiarum Fennicce, vin, 2, 395, 1866).
But Koskinnen accepted other identifications of the Ydumaei incon-
Appendix 251
sistent with a name beginning with /. Thus he supposed Ydumaea
the same as Wulfstan's Witland (Alfred's Orosius, ed. Sweet, p. 20) and
as the modern Lettish and Livonian names for Livonia, Widsemme and
Widumaa.
Accepting these last identifications of Ydumaea with Witland, etc.,
Miillenhoff pointed out that Henry's Ydumaea could in that case be only
a Latinized corruption : that the name must have begun with a PP, which
Henry had dropped in order to bring the name into line with the
Scriptural Idumaea ; and that these [Wjidumaei could therefore have
nothing to do with the Idumingas of Widsith (Miillenhoff D.A. II, 347).
The O.E. Idumingas Miillenhoff regarded as simply the biblical
reminiscence of a learned interpolator (Z.f.d.A. xi, 291 ; Grein-Wulcker,
I, 401).
MullenhofFs argument is convincing, but the premises accepted by
him and by Koskinnen are probably incorrect.
The identification of Ydumaea with Wulfstan's Witland is locally and
racially more than doubtful, and, even if Ydumaea is identical with the
modern Widseme, the W in the modern word is apparently not original,
and due either to corruption or false analogy. (See A. Bielenstein Die
Grenzen des Lettischen Volksstammes, St Petersburg, 1892, 71-74, and a
note by Prof. Vilhelm Thomson in the Magazin der Lettisch. Litterdrischen
Oesellschaft, xix, 3, 151 (1894) ; but compare also Bielenstein, pp. 466-470,
and especially the arguments of E. Kunik in favour of Miillenhoff's theory
and the originality of the W.)
Bielenstein supposes the original form of Henry's Idumaea to have
been Yduma ; and in adopting this form he is not led by any desire to
bring the word into line with the Old English, which he does not quote,
but only by the etymology of the word, which he takes to be of Livonian
origin signifying the north-east-land. Henry's corruption is therefore
confined to the second half of the word, which is obviously corrupt also in
the O.E. Iduming (cf. Lidwicing], whilst the first syllable of the word is
identical in each case. Phonetically then the identification of Idumingas
with Ydumaei is possible enough. But it cannot be accepted as certain, in
view of the great lapse of time between Widsith and Henry's Chronicle.
This difficulty is somewhat increased by the fact that the Ydumaei are a
tiny tribe occupying a territory about the size of one of the smaller
English counties. We should hardly expect so small a people to
maintain itself throughout seven centuries in such vast regions. The
identification of Henry's Ydumaei with the Idumingas of our text can,
then, be accepted only with considerable reservation.
Ydumaea is mentioned in sufficient detail by Henry to enable us to
locate it accurately (see Wiedemann, p. xxiv : A. Bielenstein's Atlas der
ethnologischen Geographie des Lettenlandes, St Petersburg, 1892). It was
Widsith
the smallest of the districts into which Livonia was divided, lying between
the Raupa (modern Brasle) and the Aa. The Fdumaei were in Henry's
day Livonians, with a sprinkling of Letts. The collocation of Idumingas
with the Iste (who are most probably Letts and Lithuanians) and of
both with the most eastern German tribes of the border, the Heruli and
Thuringians, certainly looks like a piece of genuine tradition accurately
preserved.
(H) THE TEEM HR^DAS APPLIED TO THE GOTHS.
This term is found in O.E. and in Scandinavian sources, but not else
where. The Scandinavian documents are confused. The Icelandic
geographers of the 14th century explain Reiftgotaland as a district
bordering upon Poland (Antiq. russes, n, 438, 447). Snorri equates it
with Jutland. He evidently had a false etymology in mind, connecting
the name with reift, " carriage," for he makes it apply to the mainland, as
opposed to Eygotaland, the islands (see Sn. Edda, udg. J6nsson, pp. 7,
143) : Reffigotaland is referred to also in the Heimskringla (udg. J6nsson,
1893, i, 18). Again, in one of the redactions of the Hervarar Saga, Re&$-
gotaland is explained as Jutland (Fornaldar Sogur, ed. Rafn, I, 526).
Other Icelandic interpretations will be found in Heinzel (H&rvararsaga
in W.S.B. cxiv, 56). It is to be noted that in the fragmentary lay upon
which this portion of the Hervarar Saga is founded, the land of the Goths
lies north and west of the land of the Huns, with Mirkwood between.
This is the geography of Widsith : Goths and Huns facing, with the
Vistula wood between.
The origin of the name is equally uncertain : the Old English poets
certainly understood it to mean " the glorious Goths " (hreft, glory) and
this etymology has been supported with more or less hesitation by Rieger
(Lesebuch, 286) ; Vigfusson in Cleasby's Icelandic Dictionary, sub voce
hrbftr; Kunik in the Memoires de I'Acade'mie des Sciences de St Petersbourg
vil, xxiii, 1, pp. 381-4 ; Kern in Cosijn's Taalkundige Bijdragen, Haarlem,
1877, p. 29 ; Bugge (contrary to his earlier view) in P.B.JB. xxiv, 445-6.
Kunik, Kern, and Bugge devise different methods of bringing O.E.
hreft (from *hrdfyi) and Icel. hrety (from *hratyi) from one and the
same root.
The exact forms preserved are Reiftgotaland in Icelandic, both prose
and verse ; Hratymarar in the 10th century Swedish Runic inscription at
Rok, referring to Theodoric's realm ; Hreftcyninges, Hreftgotum, Hrceda
(gen. from a nom. Hrcedas or Hrcede) in Widsith ; Hreftgotan, HreSa,
in the Elene. (For the shortened forms cf. Wederas for Wedergeatas in
Beowulf.} Ettmiiller (T) was the first to point out that the metre demands
an h in the one passage where the word occurs in Icelandic verse, and that
Appendix 253
RetiS is therefore a corruption of #m'$ ; and this is confirmed by the h in
Swedish and O.E. Ettmuller also pointed out that the Icelandic ei does
not correspond to the O.E. e (Hre%-\ but to ce (Hrceda). This was further
emphasized by Miillenhoff (Z.f.d.A. xn, 260). The fact that one of the
forms in Widsith, Hrceda, supports the Scandinavian form (H)rei'S leads
us to regard this as original, and the O.E. Hrefi as due to false analogy.
The true derivation of ffrced (from ffrai]n) remains still to seek. Heinzel
(Ostgotische Heldensage in W.S.B. cxix, pp. 26-8) derives from Radagais
(O.E. Rcedgota} who was cut to pieces, with his Goths, in Italy in 405.
The name would then mean "the Goths of Radagais." Gering (Z.f.d.Ph.
xxvi, 26) suggests a connection with hrty, "tempest," "The storm- or
battle-Goths." Much (Z.f.d.A. xxxix, 52) would connect with hrains
pure, comparing Beorhtdene in Beowulf. None of these suggestions, nor
others hazarded by Fbrstemann in Kuhn's Zeitschrift fur vergleich. Sprach-
forschung, xix, 367, are probable.
See also Grimm, G.D.S., 446, 740-1 ; Jessen in Z.f.d.Ph. in, 72 ;
Munch, Saml. Afhandl. I, 439, n, 105-10 ; Bugge, Ueber die Insch. des
Roksteins ; Bugge in P.B.B. xn, 6, 7 ; Jiriczek (2) 127 ; Schiitte in A.f.n.F.
xxi, 37.
(I) ERMANARIC AS THE FOE OF THE HUNS.
It has recently been argued by Heusler (Berl. Sitzungsberichte,
xxxvn, 925, 1909 ; Hoops Reallexikon, Attila, p. 138) that the story of
Ermanaric as the typical foe of the Huns had no continued existence, but
was first forgotten, and then invented again about the ninth century
when, since Theodoric was represented as taking refuge with Attila,
hostile relations between Ermanaric and Attila were imagined.
The historic fact of antagonism between the Huns and Ermanaric is
recorded in the fourth century by Ammianus. It would be a strange
coincidence had the poets in the ninth century reproduced this
situation accidentally, and such coincidence can hardly be assumed with
out evidence.
But the evidence is all to the contrary. Jordanes in the sixth century
represents Ermanaric as the foe of the Huns. In Widsith the grouping of
Attila and Ermanaric
weold Hunum, Eormanric Gotum (1. 18)
might reasonably be supposed to be due to a reminiscence of their
connection in story, precisely as in the exactly parallel Icelandic lines
the contending parties are grouped together :
Ar kv6fto Humla Hiinom raSa,
Gitzor Gautom, Gotom Anganty....
Widsith
This of course is only an inference, but on the other hand it hardly
admits of dispute that in Widsith (11. 109-122) the innweorud Earmanrices
is represented as combating the people of Attila. This is certainly earlier
than the tradition which makes Attila the refuge of Theodoric; yet
Widsith was popular enough to be still transcribed c. 1000, by which
time Attila was widely known as Theodoric's protector. So that we
have an exceptionally good chain of evidence for the continuity of the
tradition of hostility between Ermanaric and Attila. I have accordingly
assumed this continuity as proved (pp. 45, etc.).
(K) THE LAST ENGLISH ALLUSION TO WUDGA
AND KAMA.
The following passage, discovered by Dr Imelmann in MS. Cotton
Vesp. D. iv (fol. 139 b), is of the utmost importance for the history of
heroic tradition in England :
In diebws illis imperante Valentiniano imperatore uel pn'ncipe,
regnum barbaroram et germ&norum exortum est. Surgentesqwe
popwli et naciones per totam europam cowsederuwt. Hoc testantw
gesta rudolphi et hunlapi, Unwini et Widie, horsi et hengisti, Waltef
et hame quorwm quidam in Italia, quidam in Gallia, alii in bn'tanwia
ceteri uero in Germania armis et rebus bellicis claruerwit.
The MS. is late : but whatever be the source of the information, it
seems clear that the stories of Wudga, Hama, Hrothwulf, Unwen,
Hunlaf(ing), Hengest and Hors(a) were current till so long after the
Norman conquest that it was possible for these heroes to be classed with
Waltef. Dr Imelmann is to be congratulated on an important discovery,
which compels us to alter our ideas as to the date when these stories may
have become extinct. See D.L.Z. xxx, p. 999 (April 1909).
(L) THE GEOGRAPHY OF THE OROSIUS COMPARED WITH
THAT OF WIDSITH.
1. The statement of Widsith that the realm which Offa won was the
greatest of kingdoms is confirmed by the addition to the Orosius (ed.
Sweet, p. 19). However we read this passage, it can hardly be interpreted
otherwise than that the old Anglian home extended two days' sail (i.e.
between one and two hundred miles) from its centre in Schleswig, and
Appendix 255
embraced "many islands1." Apparently SUlende and Gotland are also
included in the old home of the Engle. SUlende may be Silund2 (Zealand),
but it is possibly Sinlendi, a district in the south of Jutland3. Gotland is
almost certainly Jutland (Reift-aotaland). It is not quite the case (as
Bjorkman argues, Engl. Stud, xxxix, 360) that there is no philological
difficulty in taking Gotland as Jutland : Ohthere would have said Jotland,
and this should have been written Geotland, not Gotland: but seeing
that this portion of the Orosius is extant only in one late MS., which is
full of misspellings, this error presents no insuperable difficulty. It looks
then as if the cynerica mcest of Offa might well have included the greater
part of the present Denmark, as well as Schleswig.
2. Only the larger tribes are mentioned in the Alfredian description
of Germany : there is nothing like the exclusive interest shown in Widsith
in the tribes dwelling on the Baltic, though the Baltic islands are carefully
enumerated. The difference is between the current knowledge of a well
informed man in the ninth century and a geographical knowledge, more
detailed within its narrower sphere, going back to earlier times.
3. Brandl has suggested that the " Biblical interpolation " is derived
from the Orosius. But the coincidences are not very numerous, and the
forms used differ. Contrast Moidum with MctfSum, MeSum ; Istum with
Estum, Osti.
(M) SCHtfTTE'S LAW OF INITIAL AND TERMINAL STRESS.
Oldsagn om Godtjod, by Dr Gudmund Schutte, has been frequently
quoted above with reference to details : one general point needs a fuller
examination. Schutte formulates the law that, in Germanic name-lists,
the first member is the one of greatest general importance; the last the
one in which the framers of the tradition have the most special interest.
Dr Axel Olrik has emphasized the value of this law as "a test in cases
which have hitherto perplexed the student" (Folk-Lore^ xix, 353, 1908),
1 See above, p. 72. In favour of the interpretation of Denemearc there
given, it should be noted that Adam of Bremen (in describing that same portion
of the present southern Sweden, which Ohthere is coasting) makes the Danish
territory stretch up to Norway : primi occurrunt Nortmanni, deinde Sconia
prominet, regio Danorum (Adam of Bremen in Pertz (fol.) vn, 373). Adam
regards Sconia as an integral part of Denmark: Sconia est pulcherrima visu
Danice provintia. To Adam the boundary between Denmark and Sweden is
not, as at present, the sea, but the forests and uplands of what is now the
interior of Sweden : Sconia est pars ultima Daniee fere insula, undique enim
cinr.ta est mart prater unum terra brachium quod ab oriente continens Sueoniam
disterminat a Dania. This is exactly the geography of Ohthere.
2 For the form and interpretation of Silund, cf. Much in P.B.B. xvn, 196;
Kossinna in I.F . vn, 281.
3 Chadwick (104, note); cf. Schiick (19) " Sonderjylland." It can hardly
be, as Sweet thinks, Holstein.
256 Widsith
and suggests that "Schiitte's Law" may prove as efficient in folk-lore
analysis as has Verner's Law in phonetic analysis.
That the law represents a widely-spread tendency of the mind seems
clear : it will be found to apply very generally. I take the first two
lists which occur to me. The list of David's mighty men of valour
(II Samuel, xxiii) begins with Adino the Eznite, chief of the captains and
hero of the greatest achievement ("he lift up his spear against eight
hundred, whom he slew at one time ") — clearly the one of greatest general
importance. It ends with Uriah the Hittite — certainly the one of greatest
special interest. So in Mallory's Morte & Arthur: where the "hundred
knights and ten " of the Round Table are enumerated, the first to search
Sir Urre's wound is King Arthur, the last Sir Launcelot.
But I doubt whether in Widsith any such law is consciously operating.
It is claimed that the Catalogue of Kings begins with Attila and ends
with Offa (Olrik, p. 353). But is Attila, rather than Ermanaric, the king
of greatest general importance 1 And the list does not end with Offa.
If we include the supplementary lines (35-49), it ends with Hrothwulf,
Hrothgar and Ingeld : if we omit these lines, it ends with Alewih. The
list of Gothic champions begins with Hethca, otherwise unknown. It
ends with Wudga and Hama, who certainly seem to be of most special
interest to the poet. But that he did not consciously regard the last
place as due to those in whom he had a special interest is shown by his
apology for so placing them :
ne wseran J>aet gesij>a >a saemestan,
)>eahj>e ic hy anihst nemnan sceolde.
(N) ON THE BEAG GIVEN BY EORMANRIC.
The beag is said to contain refined gold to the value of siex hund
sceatta scillingrime. The stilling (not an actual coin, but a denomination
of money value) was equal, in Mercia, to four silver sceattas. The sceatt
was a current coin, but the term is also used generally in the sense of
" money," and I take that to be its force here, " money to the quantity
of six hundred counted in shillings," i.e. the beag is worth 600 shillings
( = 2400 sceattas, using sceatt in its narrower sense).
This rendering seems preferable to the alternative one, of taking sceatt
strictly and stilling vaguely, "six hundred sceattas reckoned in money,"
because :
(1) It is easy to parallel the use of sceatt as equivalent to "money"
generally, indeed this is its normal force, but it would be difficult so to
explain stilling.
(2) If we take the Anglian reckoning of four silver sceattas to the
shilling (Seebohm, 363) and allow twenty sceattas to the Roman ounce
Appendix 257
(Seebohm, 443—455) and reckon gold to silver as 1 : 12 (Seebohm, 451) we
find that a golden beag worth 600 sceattas would have weighed 2£ ounces,
one worth 600 shillings ten ounces. Since the gift is mentioned as note
worthy, the latter interpretation is distinctly preferable, in view of the
considerable weight of existing gold necklets and armlets. Cf. e.g.
Odobesco (A.), Le Tremor de Petrossa, Paris, 1900 ; Rygh, NorsJce Oldsager,
Christiania, 1885 (especially Nos. 704, 705, 707, 708, 714 a); Archaeologia,
xxxm, 347 ; xxxix, 505, etc,
On the other hand, to take the beag (with Holthausen) as worth
600 gold pieces would be to make it altogether too massive, and seerns
hardly possible linguistically, for though the sceatt is, exceptionally, a gold
coin, the stilling is, as stated above, a denomination for the value of so
many silver sceattas.
c. 17
10
The dotted lines are drawn
to connect different localities
inhabited at different times
by one tribe. No attempt
is made to indicate the exact
route taken by that tribe
la fts migrations.
Long. 10 E. of Green w. 15
8 SciringesheaL
12
The dotted line represents
the course apparently
followed by Obthere.
54
INDEX
^gelmund, 121, 220
JElfwine, see Alboin
JSnenas, 209
2Estii, 248 etc.
Mtla, 191, 222. See also Attila
Agathias, his account of the death of
Phulcaris, 60
Alboin (O.E. jElfwine), 123 etc., 211
Alewih, 203
Alexander the Great, legend of, 190-1
Alfred the Great, his interest in
"Saxon" songs, 2; on the original
home of the Angles, 71-2 note, 241
etc.
Ammianus Marcellinus, his account of
Ermanaric, 16
Ammius (Hamido, O.N. Hamthir), 17,
19, 29
Amothingas, 215
Angel, 71-5, 189, 241 etc.
Angles, Angli, see Engle
"Anglo-Frisian," 154
Anglo-Saxon Chronicle, the genealogies
in, 119 etc.
Angul, brother of Dan, 76-7
Athislus, see Eadgils
Attila (O.E. Mil&; M.H.G. Etzel;
O.N. Atli), 44-6
Audoin (O.E. Eadwine), 122 etc., 212,
220
Authun the Icelander, and Harold
Hardrada, 25-6
Aviones, see Eowas
Baningas, 191
Bardengau, 116
Beadeca, 218, 219
Beanstan, 110 note
Becca, Bicca (O.N. Bikki), 19-20, 23,
191 219
Beowulf the Geat, 25, 83-4, 184
Beowulf, the story of Sceaf in, 118
Bikki, see Becca
Billing, 196
Bojunga, on the date of Widsith, 34
Bragi the old, his Eagnars drdpa, 19-
20 note; 101
Brandl, Prof. Alois, on Widsith, 140-
1,212
Breca, 110-11
Brondingas, 111 note
Burgundians, 58, 161-2 note, 192, 210
Gaelic, 192
Casere, 192, 212
Cassiodorus, his Variae referred to, 13,
14, 35, 37 note
Chadwick, Mr H. M., on Widsith, 141
etc. ; on Germanic ethnography, 156
note
Charles the Great, his interest in the
ancient songs, 2
Chauci, 67-9
Creacas (the Greeks), 166, 192, 212
Dan, king of the Danes, 76
Danes, 75 etc.
Daukiones, 75 note
Deanas, 210
Dear's Song, 4, 40, 42, 105
Dialect of Widsith, 166
Dunheath, battle of, 46-8
Eadgils (Saxo, Athislus; O.N. Athils),
92-4, 217
Eadwine, see Audoin
Ealhhild, 21-8, 217
Eastgota, 13-15, 219, 236-7
East-Thyringas, 215, 216
Eatul (Italy), 211
Ebreas, 215
Egyptas, 215
Elsa, 122, 220
Emerca, see Harlungs
Engle (Angli), 69, 73, 209, 245 etc.
Eolas, 8, 216
Eonnanrie, see Ermanaric
Eotas, see Yte
Eowas (Aviones), 69, 73, 197
Ermanaric (O.E. Eormanric; M.H.Q.
Index
261
Ermenrich, Ermrich; Thid.saga, Er-
minrekr ; O.N. Jormunrekkr ; Saxo,
larmericus), 15-40, 48-55, 191, 216
etc., 253
Erminrek, see Ermanaric
Ethel werd, on the Sceaf story, 118
Eudoses, 69, 73-4
Exsyringas, 215
Fifeldor, 204
Finn Folcwalding, 67
Finnas, 192, 212
Franks, 112 etc. See also Froncan
Freotheric, 222
Fresna cynn, see Frisians
Fridla, see Harlungs
Frisians, 67, 197, 211
Froda, 79
Froncan, 195, 211
Frumtingas, 68
Frysan, see Frisians
Gado, see Wade
Geatas, 207
Gefflegas, 209
Gefwulf, 197
Gensimund, 183
Gepidae, 15, 209
Gifica (M.H.G. Gibeche ; O.N. Giuki),
64, 191
Gifthas, see Gepidae
Gislhere (Gislahari), 65, 222
Gizur, 47
Glommas, 108-9 note, 193, 211
Gotland, 254-5
Gregory the Great, his opinion of the
Lombards, 116
Gummere, on the date of Widsith, 150
Gundahari, see Guthhere
Gunnar, see Guthhere
Gunther, see Guthhere
Guthhere (Gundahari; M.H.G. Gun
ther; O.N. Gunnarr), 58-63, 211
eelsingas, 108, 194
Hserethas, 214
Huethnas, 214
Hastwere, 201-2
Hagena^.-ff.G. Hagen; O.N. Hogni;
Saxo, Htfginus), 100 etc.
Halga (Saxo, Helgo; O.N. Helgi), 82
Hama (O.H.G. Heimo; M.H.G. Heime;
O.N. Heimir), 52-7, 222-3, 254, 256
Harlungs, story of the, 28-36, 218-19
Heatho-Beardan, 79-81, 82, 205
Heatho-Eeamas, 210
Heathoric, 219
Hebrews, see Ebreas
Heimo, Heime, Heimir, see Hama
Heinzel," Richard, on Ealhhild, 22,
132 ; on the Hervarar saga, 47-8
Heligoland, 71
Helm, 198
Hengest, 254
Heoden (M.H.G. Hetel; O.N. Hethinn;
Saxo, Hithinus), 100 etc., 193
Heorot, 78 etc., 205
Heoroweard (Saxo, Hiarwarus), 84
Herefaran, 202
Herelingas, see Harlungs
Herminones, 155 etc.
Heruli, 216, 245
Hervarar saga, 46-8
Hethca, 218
Hild, 101 etc.
Hildebrand, master, 40, 52
Hlithe (O.N. Hlothr), 46-8, 219
Hnarf, 197
Hocingas, 68 note, 197
Hodbroddus, 81
Holen, 202
Holmryge, see Rugians
Horsa, 254
Hreedas, 163, 221, 252
Hrethcyning, 189, 252 etc.
Hrethric (Saxo, Rtfricus), 83, 84 note
Hringweald, 202
Hronas, 210
Hrothgar (Saxo, Roe; O.N. Hroarr),
81-3
Hrothwulf (Saxo, Roluo; O.N. Hrolfr),
81-4, 254
Hugas, 68 note
Hug-Dietrich, see Theodoric the Frank
Humli, 46-8
Hun, 201
Hundingas, 195, 214
Hungar (perhaps Hunigaisus, Onege-
sius), 220-1
Hunlaf(ing), 254
Hunuil, see Unwen
Hwala, 190
Hydatius, 62
Idumingas, 8, 216, 250 etc.
Incgentheow(O..N. Angantyr), 46-8, 219
Indeas, 215
Ing, 77 note
Ingaevones, 155 etc.
Ingeld (Ingellus, Hinieldus), 78 note,
79-81
Ingwine, 77 note
Iring, 114, 115
Irminfrid, 114, 185
Island-Rugians, see Rugians
Israhelas, 215
Istaevones, 155 etc.
Iste, 8, 216, 248 etc.
Jordanes, his account of Germanic
songs, 13-14 ; of Ostrogotha, 14-15 ;
of Ermanaric, 16-19, 53
262
Index
Jormunrekk, see Ermanaric
Julian the Apostate, on Germanic
song, 2
Jutae, see Yte
Ket (Saxo, Keto), 85, 93-4
Kudrun, 102
Kynaston, Sir Francis, on Wade, 97
Langobardi, see Lombards
Lawrence, Dr W. W., on the structure
of Widsith, 136 etc.
leben in place names, 245
Leonas, 213, 214
Leth(uc), a Lombard king, 122
Lex Angliorum et Werinorum hoc est
Thuringorum, 245
Lichtenheld's tests, 167 etc.
Lidwicingas, 213
Lombards, 115 etc., 201, 205-6, 213
Lombard and Gothic story perhaps
connected, 23, 122
Longbeardan, see Lombards
"Low German," ambiguity of the
term, 154
Loyalty as a motive in Germanic lite
rature, 59-60; in Germanic life, 183
etc.
Malmesbury, William of, on the
"Sceaf" story, 118
Map, Walter, his tale of Gado, 99
Maurer, K., his views on the date of
Widsith, 7, 147
Mauringa, 161
Maurungani, 159-61, 235-6
Meaca, 195
Mearchealf, 195
Merseburg Glosses, 242
Moller, on the structure and date of
Widsith, 130 etc., 150
Mofdingas, 215
Moidas, 215
Morris, William, on the Sigfrid story,
61,63
Morsbach's tests, 171
Mullenhoff, Carl, on the story of the
Harlungs, 30-3 ; on Gundahari, 62 ;
on the Offa-story, 90 ; on the struc
ture and date of Widsith, 128 etc.,
150
Myrgingas, 159 etc., 195, 204, 215, 217
Nerthus, 69-71
Nibelungen Lied, 3, 61
Njorth, 70-1
Nuithones, 69, 74
Nuodunc (Thid. saga, Nauthung), 51
Odoacer, 29, 30 note, 39-40, 59
Offa, king of Angel (Saxo, Uffo), 84-92,
202
Olrik, Dr Axel, on Widsith, 147, 150
Onegesius, see Hungar
Ongel, see Angel
Ongentheow, 200
Orosius, 71
Ostrogotha, see Eastgota
Oswine, 197
Panzer, on the story of Hagena and
Heoden, 105-8
Paul the Deacon, on the death of Gun
dahari, 62-3; on Agelmund, 121,
123; on Alboin and Thurisind, 123
etc. ; on Mauringa, 161
Peohtas, 213
Persas, 215
Phulcaris, death of, 60
Priscus, his account of minstrelsy at
Attila's court, 175
Prosper Aquitanus, 62
Ptolemy's Geography, 243 etc.
Baedhere, 222
Eavenna, Geographer of, 159-61
Keudigni, 69, 73
Eichter, Carl, 171 etc.
Eondingas, 196, 222
Eugas, see Rugians
Eugians, 192-3, 211
Eurasian, 222
Eumwalas, 211
Sffiferth, 199
Sarus (Sarilo, O.N. Sorli), 17, 19, 29
Saxo Grammaticus, on the Swanhild
story, 19-20; on Dan and Angul, 76
-7 ; on Ingellus, 80 ; on Offa, 85-8 ;
on Hithinus and Eteginus, 103
Saxons, 68-9, 73, 210
Sceafa, 117, 201
Sceafthere, 200
sceatt, 217, 256
Scilling, 218
settling, 256
Scottas, 213
Scride-Finnas, 213
Seafola (M.H.G. Sabene), 41, 219
Seaxe, see Saxons
Secca, 219
Sercingas, 212
Seringas, 212
Sibich, see Sifeca
Sidonius Apollinaris, on Germanic
song, 2; on the defeat of Gundahari,
62
Siebs, Prof. Theodor, on Widsith, 143
etc.
Sifeca (M.H.G. Sibich; Thid. saga,
Sifka), 32-3, 51, 219
Index
263
Sigehere, 197
Sillende, 254-5
Snorri, his story of Hogni and Hethin,
101
Sorli, see Sarus
Speght, on Wade, 96
Starkath, 81
Suardones, 69, 74
Sueboi Aggeiloi of Ptolemy, 243
Sunilda, see Swanhild
Swaefe, 194, 204-5, 209
Swanhild (Sunilda), 17-19
Swedes, 199
Sweordweras, 210
Sweyn Aageson, on Offa, 85-8
Sycgan, 199, 210
Symons, on the date of Widsith, 150
Tacitus, on Germanic song, 12; on
Germanic loyalty, 59; on the Ger
manic coast tribes, 66-7; on the
Nerthus-worshippers, 69-75; on the
ethnographical divisions of Ger
many, 155
Ten Brink, on the structure and date
of Widsith, IBletc., 150
Theodoric of Verona (M.H.G. Dietrich
von Bern; Thid. saga, Thithrekr),
13, 36-44, 51, 54, 219
Theodoric the Frank, 112 etc., 185,
195
Thiadric, see Theodoric the Frank
Thidreks saga, quoted passim, but see
especially 61, 69
Throwendas, 210
Thuringia, 245 etc. See also Thyrin-
gas
Thuringian war, 113 etc.
Thurisind and Thurismod, 124
Thyringas, 198, 210. See also Thurin
gian war
Unwen (Hunuil), 219, 254
Vandili, 156
Varini, see Wernas
Vidigoia, see Wudga
Vistula-wood, see Wistla wudu
Vitae Duorum Off arum, 88-90
Vithga, see Wudga
Vylsunga saga, 61
Wade (O.E. Wada; M.H.G. Wate;
Thid. saga, Vathe), 95 etc., 102-3
Wsernas (Varini), 69, 73, 196, 245 etc.
Wala rice, 212
Wald, 198
Waldere fragment, 42
Warni, see Varini
Wends, 209
Wenlas, 208
Wermund, 85-8
Wernas, see Warnas
Wicinga cynn, Wicingas, 205, 208
Widia, see Wudga
Widsith, 188
Widukind, 113 etc.
Wig (Saxo, Wigo), 85, 93-4
Winedas, see Wends
Wistla wudu, 142, 163, 221-2
Witege, see Wudga
Withergield, 222
With-Myrgingas, 221
Witta, 194
Wod, 198
Woingas, 198
Wolf -Dietrich, 112 etc.
Worms, Burgundians at, 58
Wrosnas, 202
Wudga, Widia (Jordanes, Vidigoia;
M.H.G. Witege; Thid. saga, Vithga),
42-3, 48-53, 55, 222-3, 254, 256
Wulfhere, 221
Wulfingas, 198
Wynnhere (O.N. Ormarr), 47, 221
Ydumaei, 250 etc.
Ymbre, 200
Yte (Jutae), 73-4, 237-41
CAMBRIDGE I HUNTED BY JOHN CLAY, M.A. AT THE DNIVEB8ITY PRESS.
WING SZrr. NOV 6*72
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
PR
1788
05
1912
Chambers, Raymond Wilson
Widsith
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11