Introduction To The Older Runic Inscription

Roland Schuhmann JSIEL 2009 The early Runic Inscriptions
Lehrstuhl fr Indogermanistik
Friedrich-Schiller-Universitt Jena
Jena Summerschool of Indo-
European Languages 2009
NTRODUCTION
Runology = scientific occupation with the ancient documents
furnished with indigenous germanic characters.
Rune = now only used in the sense letter or character of the
earliest Germanic alphabet is a scientific loan in the 17
th

century form danish rune letter or character of the earliest
Germanic alphabet (cp. also swedish runa id., norsk
dialectal runa old form).
The word rune is passed down in the earliest inscriptions
several times:
stone of Einang (ca. 350-400): acc.sg. runo;
stone of Noleby (ca. 450-600), brooch of Eikeland (ca.
450), clasp of Freilaubersheim (ca. 575): acc.sg. runa;
stone of Stentoften (ca. 600-650), stone of Bjrketorp
(ca. 600-650): gen.pl. runo;
stone of Stentoften (ca. 600-650): acc.pl. runoz; stone of
Bjrketorp (ca. 600-650): acc.pl. runAz; stone of Jrsberg
(ca. 450), bracteate of Tjurk 1 (ca. 500), stone of Istaby
(ca. 600-650): acc.pl. ronoz;
stone of Bjrketorp: dat.sg. ronu.
Outside the runic inscriptions the word is found latinised in a
poem of Venantius Fortunatus (carm. 7,18,19-20; end of the
6
th
century): barbara fraxineis pingantur rhuna tabellis:
quodque papyrus agit, virgula plana valet the foreign rune
may be painted on ashen tablets, what is done by papyrus,
can also be done by a smooth peace of wood.

The Germanic runes are to be seperated from symbolic
signs (sometimes semi-identical) that occur since the
Bronze Age in the Germanic area as well as from the so
calles Hunnic and Turkish runes.
The Greek and Latin characters are unlike the runic
characters named with the word protogerm. *ka-staa-
beech-stave > ohg. buohstab, osax. bkstaf, oe. bcstf,
oisl. bkstafr character, alphabet.
The etymology of the word rune is disputed. Two different
proposals compete against another:
Most widely accepted is the connection with a wordgroup
that is found in every germanic language: goth. runa
, , , secret, decision,
consultancy, ohg., osax. rna murmur, secret, whisper,
oe. rn secret, counsel, consultancy, oisl. (pl.) rnar
secret < protogerm. *rn- secret, whisper. A verb
derived from this substantive is found as: ohg., osax.
rnn to whisper secretly, oe. rnian to whisper, oisl.
rna to have an intimate conversation; to execute runic
magic.
Outside Germanic exact equivalents are only found in the
Celtic languages: oldirish rn, cymrian rhin secret. But
the precise relationship between the words in both
language families is unclear: inerited words from indo-
european or loanwords (when this is the case: in which
direction?). This etymology is considered to be an
important piece of evidence for the magic character of the
runes. Further away stand: gr. to ask, lat. rmor
noise, lith. runt to speak.
Less accepted is the assumption of two homonymous
roots protogerm. *rn-. Out of protogerm. *rn-
1
arose
the wordgroup of goth. runa, out of protogerm. *rn-
2
the
word rune. *rn-
2
would have as continuants mlg. rune
gelding, mld. ruun, ruyn, rne gelding, mld. rnen,
ruynen caponize, nfris. rn gelding, nfris. runje
caponize, nswed. runa recess, groove, rill. It would be
an extension with *-n- of the protoie. root *(H)re H- to
rip, to drag, to dig > ocsl. ryj to dig, rylo spade, lith.
ruti to dig out, ruobti to scratch, to cut, mir. ruamar
to dig out, ruam spade, lat. rutrum spade.
Protogerm. *rn-
2
would have the meaning the cut, the
incision, so that it would be the term for the technique,
becoming extinct in the southern Germanic area with the
disappearing of the runic knowledge but being
continuated in Scandinavia. This etymology is supported
by the fact that the word rune in the runic inscriptions has
never something to do with magic, but only denotes the
written inscription.
o Not to the word rune belongs the name of the germ.
seeress Aurinia. The name is often, but wrongly
emended to Albruna.
All runic inscriptions are written with runic characters, but not
in the same character style. The inscriptions from the
beginning of the transmission till the end of the 7
th
century
are written in type consisting of 24 characters, the so called
elder fuark. These inscriptions constitute only a small
portion of the total records, namely some 400. The elder
fuark is ca. 700 replaced in Scandinavia by a more
simplified runic row, the younger furk, containing 16
characters resolving in some subcategories. In the Anglo-
Frisian area emerges also a modified runic row, here
increased up to 33 runes, the so called fuorc.

The runic characters are primarly used to render the respective
Germanic language. Only from the 11
th
century (after the
Christianization) the runes are also used for writing Latin
texts.

The total of known runic inscriptions lies currently about 6900.
From thes ca. 4000 fall upon Sweden, ca. 1600 upon
Norway, ca. 850 upon Denmark, ca. 80 upon Germany, ca.
90 upon Britain, ca. 25 upon the Netherlands, ca. 100 upon
Iceland, ca. 100 upon Greenland, ca. 50 upon the Orkneys,
ca. 10 upon the Faroe Islands and cq. 20 upon Ireland.
The exact numerical data differ in the literarure a little bit
because for example the number of fakings is disputed. This
listing doesnt take account of small numbers of runic
inscriptions in some regions like the inscriptions from Russia
which are summed under the Swedish inscriptions.
Temporally the runic inscriptions range from the 2
nd
century to
the 19th century (inscriptions from the Swedish region of
Dalarna; these are to be linked to a scientific tradition, so
they cannot be regardes as popular). In Scandinavia the runes
were untill the 15
th
century in actual use.
The areal borders of the findings range in the north till the
island Kingigtorssuaq offshore Greenland (a small runic
stone) and till Senia near Troms (a necklace); in the south
till Piraeus in Greece (a marmoreal lion); in the west till
Ireland; in the east till Staraya Ladoga in Russia (a wooden
stick) and the Island Bereza in the orifice of the Dnjepr (a
memorial stone).
The number of runic inscriptions increases continually by
means of archaeological findings. These new findings are
published since 1986 in the Journal Nytt om runer
(http://www.ukm.uio.no/runenews/).
The runic inscriptions found in America have untill now all
been shown to be fakings.
The runic script is primary used as an epigraphical script. So it is
applied to solid materials. As is the case with other epigraphical
documents the runic inscriptions cann be divided in two groups:
Inscriptions on loose objects like weapons, clasps, coins,
objects of daily use;
Inscriptions on stones and stone-plates.
The difference between both groups is fundamental: As stones and
stone-plates are in principle immobile objects, the loose objects
are mobile and can be transported over immense distances. The
origin of a runic inscriptions on mobile objects must not be
identical with the place of origin of the object and must not be in
accord with the context of the finding.
HE ELDER FUARK
On some objects from the 5
th
and 6
th
century there is found
complete or partially the runic row of the elder fuark in
more or less the same order. According the first six
characters it is called fuark (cp. abc, alphabet). Ist order
differs from that of the Mediterranean alphabets.
The inscriptions reveal a standard runic row that can be
rendered as:
f u a r k g w

h n i j p z s

t b e m l d o

For the translitaration of the runic characters small bold latin
characters are used. A dot underneatha transliterated
character indicates that the reading of the rune is not secure.
The sign * marks an assumable runic character of unknown
value. The sign ? is used when it is unclear whether a rune is
existent at all or whether a scratch or a decoration sign has to
be assumed. So called bind-runes are notified by a bow (cp.
az). Broken parts of an inscription are idicated by ], [ or
[].
Most runic signs prefer in their basic form angular and vertical
shapes. This finds it reason certainly in the material of the
inscriptions, in which rounded forms were more difficult to
inscribe.
Although the single runic signs consist only out of a few components,
a formal consistent, structural analysis ist difficult. Many
proposals were put forward, but none of the has been able to
establish oneself. The following formal analysis is new. The
components, which constitute through their combination the single
runig signs, can be determined as follows:
| staff;
branch;
twig;
< crook;
zigzag.
The constituents can be defined as follows:
staff: staffs stand always upright and normally achieve
the full line height;
branch: branches are sloping staffs; they can be longer as
well as shorter than staffs; there can be discerned left-
sloping branches (\; from top left to down right) and
right-sloping branches (/; from down left to top right); in
combination with a staff a branch can intersect the staff,
in combination with two staffs the branch is in between
them;
twig: twigs are sloping elements too, but on the one hand
they are clearly shorter than branches, on the other hand
they only appear with one staff and dont intersect with
them, but always go off them; like the branches the twigs
are divided in left-sloping twigs (; from top left to down
right) and right-sloping twigs (; from down left to top
right);
crook: crooks consist of two sloping merging elements of
the same length; when combined with a staff the cone end
points normally away; crooks can be opened left (),
opened right (), opened at the top () or opened at the
bottom ();
zigzag: zigzags consist out of three superposed, equally
long elements.
o zigzags can also be defined as consisting out of two
overlapping crooks;
o this classification is not the only possible one, cp. for
example:
o Antonsen: staff, twig, pocket, crook;
o Dwel: staff (doubled staff), twig, crook (doubled
crook);
o Nowak: staff, twig, curve, spike, crossbar, joining
bar, diagonal, crook, zigzag, square, loop,
chevron, hook.
As the standard runic row shows the runes consist either only
out of a staff, a crook or a zigzag or out of a combination of
the five possible elements. There can be combined: staffs
with branches, twigs, crooks and zigzag (branches, twigs an
crooks can be positioned at different places of the staff);
branches with branches and crooks; crooks with crooks;
zigzags with zigzags. Not combined can be branches, twigs
and zigzags. Schematically the standard runic signs can be
presented as combinations as follows:
no staff one staff two staffs
i
branch 1 u n h
2 g m d
twig 1 l
2 f a
t z
crook 1 k w e
2 j p b
zigzag 1 s r
2 o
The table could be refined by indicating the exact place
where a twig or a crook is attached to the staff (above,
centred, below;
The runes d and m could also be analysed as consisting out
of two staffs and two crooks respectively;
The rune g could also be analysed as consisting out of two
crooks;
The rune t could also be analysed as consisting out of one
staff and one crook;
Problematic is in every analysis the rune o.
The forms of the runic signs as described above, are of course
only ideal forms (although in each case thes forms are
attested). In the inscriptions appear beside these ideal forms
also other aberrant forms, such as rounded variants. This
variance can have more then one cause: So can the material
of the object be responsible as well as the skill of the carver
or both. It is use that deviant forms are specially noted.
It is often assumed that the variants show a temporal
development, so that forms of runic signs can be used for an
approximate dating of inscriptions. This view didnt bear up
against a critical examination.
HE RUNIC WRITING CONVENTIONS
The runic inscriptions show some orthographic pecularities:
In many inscriptions ligatures consisting out of two or
more runes are found, normally called bind-runes. When
two runes appear as a ligature they have one common staff.
In some rare cases namely by runes that dont have a
staff the binding element is no staff, but a branch. When
more then two runes are combined there exists normally
one common staff per two runes. Because when possible
a staff is involved it is clear that the rune i never occurs
as part of a bind-rune.
The purpose of the bind-runes is not clear. Often it is
assumed that with bind-runes space or labour should be
saved. Considering the shortness of most runic inscriptions
this seems to be odd. A ornamental reason may be existent.
Double consonants (word internally and even when
belonging to two words) as well as double vowels
(belonging to two words) are usually written single.
A nasal is normally not written before an homorganic
consonant. This orthographic rule can have two causes: a
phonological one (the vowol before the nasal was spoken
as a nasalized vowel) or it was taken over from the giver-
script.
The direction of writing in the inscriptions is not yet
definite although in the single inscriptions it is fix.
Single-line inscriptions can be written from left to right as
well as from right to left (only once a change in the
direction is documented). In more-line inscriptions there
are more possibilities: So can on the one hand all lines be
written in the same direction (either from left to right or
from right to left) but on the other hand also in
boustrophedon style (either true or fals boustrophedon).
The direction of the writing is in most cases securely
indicated by those runic characters that are directional
(for example f, a, n).
The free direction of writing accounts also for the so called
Wenderunen (reversed runes) and the Sturzrunen (inverted
runes). The first ones are runes that seem to be written
against the normal direction of an inscription ( : ). The
latter ones are upside-down runes, so runes which are
turned upside-down in comparison to the other runes ( :
).
In several inscriptions the single words ar divided by
division marks. These consist out of one till five
superposed points or small strokes.
Sometimes the runic lines of an inscription are enclosed
by a margin or runic lines stand on lines.
In the opinion of some scholars there seems to exist
another orthographic rule: The sequence (C)VRC can be
replaced by (C) C and rendered by <(C)RC>.
1. THE OLDEST RUNIC INSCRIPTION: THE
COMB OF VIMOSE (150-160 AD)
The comb of Vimose is the oldest runic inscription so far
known; it was found in 1865 in the bog of Vi and is an
undamaged two-part comb (material: horns). The object is 56
x 49 mm. The runic characters have an height Hhe of ca. 13
mm and a length of 29 mm.; they are applied on the cover
plate of the comb. The inscription regards the threefold
perforation.
Because it is a mobile object the origin of the inscription cannot
be secured.
http://runer.natmus.dk//userfiles/image/ImgCache/919/rune06b_167.jpg
http://runer.natmus.dk//userfiles/image/ImgCache/919/Vimosekammentegn
.JPG
The inscription can easily be read as:
()

5

and be transliterated as:
harja ()

5
Interpretation: nom.sg. masc. an-stem, monothematic personal
name (cp. run. [gen.sg.] harijan [stone of Skng; ca. 500],
ohg. Herio 802, osax. Herio) < protogerm. *ar an-.
More widespread is the appellative protogerm. *ar a- masc. a-stem
army > run. -harjaz, lat.-germ. Harigasti (helmet B of Negau; 2.-
1. cent. B.C.), lat.-germ. personal names (C)hari(a)-, goth. harjis,
ohg. hari, heri, lgb. hari-, heri-, run.-preosax. (akk.sg.
?
) hari
(runic bone; 5. cent. A:D.), osax. heri, olfrk. heri-, here-, oe. here,
ofris. here, hiri, oisl. herr.
The n-stem can be explained in two ways:
harja is a derivation with the suffix protogerm. *-n-
(individualizing) of protogerm. *ar a- army and thus means
warrior;
harja is a shortening of a dithematic personal name with the
element protogerm. *ar a- army in the first part. Such
shortenings happen often to be converted to the n-stems.
Of course also other proposals have been made:
harja is an acc.sg. of a word meaning comb (not
documented in Germanic; closest parallel: lith. ers
hair); problem: where is the verb?
harja is to be connected with the tribal name Harii
member of the Harii; problem: why would such an
information be given?
Most plausible interpretation: warrior or Harja.
2. THE EARLY BOGFINDS
2.a. The bog of Illerup dal
2.a.1 Shield mount 1
This shield mount made of bronze has a weight of 0,020 kg
and a total length of 189 mm. The shield mount is
damaged severely and at different times. The shield
mount was layed as an offering in the bog. Neither the
origin of the object nor the place where the inscription
was made can be made up. The dating of the shield
mount is usualy given as 200 A.D. It was found 1976.
http://runer.natmus.dk//userfi
les/image/ImgCache/841/run
e03b_254.jpg
http://www.arild-hauge.com/arild-hauge/DK-MJy-85-ILLERUP.jpg
The inscription can be read as:
()

5

swarta ()

5
Remarks: The last rune lies under the runes 4-5; the reason
seems to be shortage of space. The rune 5 lacks the right
twig due to damage of the object.
Interpretation:
nom.sg. masc. n-stem, monothematic personal name (cp. run.
suarti [stonecross of Andreas II (Isle of Man)], ohg. Swarzo,
oisl. Svarti, odn., oswed. Swarte) < protogerm. *s artan-.
Doubtless a derivation with the suffix protogerm. *-n-
(individualizing) of protogerm. *s arta- adj. a-stem black
> got. swarts, ahd. swarz, as. swart, ae. sweart, afries. swart,
aisl. svartr.
The meaning seems to suggest a nickname; the motive (hair,
skin) remains unclear.
Direct continuant of protogerm. *s arta- adj. a-stem
black: nom.sg.m. *s artaz; problem: when the a was
added, why not the z?
swarta is an acc.sg. masc. a-stem of protogerm. *s arta-
(substantivated adj.) the black; problem: where is the
Verb? (interpretation: shelter the black one! is without
parallels).
Most plausible interpretation: Swarta [= the black one].

This shield mount is made of silver and has two rivet holes.
The weight is 0,024 kg., the total length 18,8 cm. It belongs
to a pomp-shield. It was found 1983.
In contrast to comparable findings in Illerup dal this shield
mount is almost intact. Both runic groups with a height of 5-
8 mm were inscribed after the final assembly of the object
and were carved by the same person. The space between the
two parts is caused by an (now no more preserved)
ornamental disc over the rivet.
The place of finding of the mobile object is surely not identical
with the place of origin. The place of origin (but not
necessary of the inscription) is traced to Southnorway or
Southsweden.
The time of the deposit can be dated to 207 A.D. (or shortly
later).

http://www.arild-
hauge.com/arild-
hauge/099_DR-
MS1995-
336B_Illerup-
mount-shield-
handle-2_MM.jpg
()

5

10
niijo tawide ()

5 10

Remarks: The runes 3 and 9 are so called Spiegelrunen
(mirror runes). Rune 13 has the shows a common variant.

Interpretation: Two words, word-boundary is identical with the
space.
niijo: nom.sg. masc./fem. a/n-stem, surely a shortening of a
dithematic personal name with the element protogerm.
*ni a/n- the jealous one
It is a derivation of the short personal names forming suffix
protogerm *- a/n- (cp. for the suffix for example frk.
Chlodeo, erul. Sindia) of protogerm. *na- jealousy >
goth. nei, ohg. nd, osax. nth, oe. n, ofris. nth, oisl. n.
Dithematic personal names with this appellative show up: lat.-
germ. Nitigis (bishop of Lugo), ohg. masc. Nithbald,
Nidperht, fem. Nitfalia, Nithildis; cp. also the short names
Nitho and dem. Nidilo.
It remains unclear wheather the name is masc. or fem., because
the ending -o is ambigue and the element appears as well in
masc. as in fem. names (the following verb can not settle this
case).

The word is a derivation of the weak verb protogerm.
*ni e/a- to hate (> ohg. nden, oe. nan, oisl. na);
problem: the semantic side; the name would mean
grudger, hater.
The word is a derivation with the individualizing suffix
protogerm. *-n- of protogerm. *ni a/- relative (>
goth. nijis oe. [pl.] nias, oisl. nir); problem: the
appellative *ni a/- relative doesnt appear as a
element in personal names otherwise.
tawide: the interpretation is undoubted and without controvery:
3.sg.ind.pret. of the verb protogerm. *ta e/a- weak verb I do,
make > goth. taujan, ohg. zouwen, mlg. touwen (cp. ohg.
zwn, oe. tawian weak verb II; oisl. tja [< *t e/a-]). The
verb shows up in the runic inscriptions repeatedly.
Most plausible interpretation: Niijo [= the jealous one] made.
It is unclear whether this inscription has to be interpreted as
naming the producer of the object or of the inscription.

This shield mount is made of silver and has two rivet holes.
One rivet is still attached. The weight is 0,036 kg., the total
length 19 cm. It belongs to a pomp-shield. It was found 1983.
http://www.arild-hauge.com/arild-hauge/100_DR-MS1995-336C_Illerup-mount-
shield-handle-3_MM.jpg
This shield mount reveals damages that cannot be attributed to a
battle, so were probably made when laid down in the bog
(ritual damaging). The runes have a height of 5-7 mm and
were carved in direction of the shield holder. The fifth rune is
damaged.
The place of finding of the mobile object is surely not identical
with the place of origin. The place of origin (but not
necessary of the inscription) is traced to Southnorway or
Southsweden.
The time of the deposit can be dated to 207 A.D. (or shortly
later).

()
5

laguewa ()

5

Remarks: The runes 5 and 7 are so called Spiegelrunen
(mirror runes). Rune 6 has a doubled branch.

Interpretation:
nom.sg. masc. n-stem, dithematic personal name, consisting of
the terms lagu- and -ewa.
lagu- continues most probably protogerm. *lau- water, sea >
osax., oe. lagu, oisl. lgr).
Less probable is the connection of lagu- with protogerm.
nom.pl. *la (sg. protogerm. *laa- neutr. > oe. -lg [in
gelg plot, area], osax. [pl.] -lagu [in aldarlagu defined
lifetime, gilagu destiny, lot], oisl. [pl.] lg law [sg. lag
situation, position]), because in a compound the first
element is usually the stemform.
-ewa continues protogerm. *e an- servant > oe. owa, a
derivation with the individualzing suffix protogerm. *-n-
from protogerm. *e a- servant > goth. ius*, ohg. dio-
(in dioheit humility, lowliness and in personal names), oe.
ow, oisl. -r, -r.
Against this interpretation as a masc. n-stem has been objected
that no masc. n-stems appear as second element in personal
names. But under the oldest germ. names there do show up
(although only few) such names: Liubigoto, Ultrogotho and
Vultrici(a), so this argument is not decisive.
The form would be a westgerm. masc. a-stem nom.sg. *e
az with dropped -z; problem: first the -a- would have been
dropped.
It would be the regular continuant of masc. a-stem nom.sg.
protogerm. *e az with a faulty omission of -z by the carver;
problem: rhe reason is not obvious (there is no problem of
space).
It would be an acc.sg. of a masc. a-stem protogerm. *e az
with the regular ending -a; problem: where is the Verb?
(interpretation: shelter laguewa! is without parallels).
Most plausible interpretation: Laguewa [= water-servant].
laguewa can directly be compared with the name erul.
swamp-servant and seems possess a religous
reference (cp. also lgb. Arintheus servant of a fireplace for
an offering). The cultic background deliver the apperantly as
as an offeringplace functioning bogs in Scandinavia.
Unclear remains if the name is that of the objectowner, of the
carver of the inscription or of the producer of the object.
2.a.4 Lanceheads 1/2

The lanceheads are undamaged pieces of the Type Vennolum
with a midrib. The importance of both identical inscriptions
lies in the fact that the inscriptions are not carved but
stamped. A third identical inscription comes from a
lancehead of the bog of Vimose. The Illerup inscriptions
were discovered 1980.
The lancehead 1 weighs 160 g and has a total length of 296 mm.
()

5
wagnijo ()

5

Remarks: The rune 1 is a so called Spiegelrune (mirror rune)
in a rounded form. Rune 6 is a so called Wenderune.
Before Spiegelrunen were regocnized this inscription was
transliterated as ojingaz ().
Interpretation:
nom.sg. masc./fem. a/n-stem, either a personal name or the
lance-name. wagnijo continues protogerm. * ani a/n- a
derivation with the suffix protogerm. *-(i) a/n- that builds
on the on hand side personal designations on the other hand
sind nomina agentis. The basis ist protogerm. * ana- masc.
a-stem vehicle > ohg., osax. wagan, oe. wgn, ofris. wein,
oisl. vagn, itself a derivate with the suffix protogerm. *-na-
(nomina actionis) of protogerm. * ee/a- to move > goth.
(part.pret.) -wigana, ohg., osax., oe. wegan, ofris. wega,
weia, oisl. vega. These derivatives can later become concrete
nouns. The basic meanig for protogerm. * ana- is therefore
movement.
The exact date for the transition of the meaning from
movement to vehicle is not clear. When it has to be dated
after the derivation wagnijo, then the meaning of wagnijo
would be mover (cp. goth. liugn lie : liugnja lier); this
would point to a weapon-name. But when the transition of
the meaning lies before the formation of wagnijo, then the
latter would mean vehicle-maker, vehicle-builder; this
would point to a personal name.
Most plausible interpretation: Wagnijo [= mover/vehicle-
maker].
2.a.5 Grip of a firesteel

The wooden grip of a firesteel in form of a
needle has a length of 90 mm und a diameter
of 21 mm. The down end is ornamented with
zigzags. The runic inscription discoverd 1992
has an height of 10 mm and was written on a
carved line. The inscription takes care off the
hole.
http://runer.natmus.dk//userfiles/image/Im
gCache/848/ildst%C3%A5l.JPG

The deposit of the object can be dated
at 207 A.D. The dating of the runic
inscription is less secure, the generall
dating is 200 A.D. The place of the
inscription is also not known because
the object is mobile. The making
place of the object can probably be
determined as Scandinavia.
http://www.arild-hauge.com/arild-
hauge/DK-MJy-93-ILLERUP.jpg
The inscription seems to be carved by an unskilled person; the
best reading is:
()

5
gauz ()

5

Remarks: especially the 5
th
rune calls for attention. It seems to
consist out of a staff and a crook. Problematically the crook
has the height of the whole staff; therefore a reading as w
would also be possible. Because a reading gauwz makes no
sense, the reading as is more attractive.
Interpretation:
nom.sg. masc. a/cons.-stem < protogerm. *a (a)- yeller, a
derivation with the verbal nouns formative suffix protogerm.
*-(a)- of protogerm. *a e/a- to yell > oe. gian,
westfris. geije, oisl. geyja (cp. Also mhg. gude joy, oe.
ga sillyness, ridicule, oisl. gau yelling, bluster [<
protogerm. *a -]).
Problem:
The ending -z is problematic. As the word is quite a late
derivative of the verb, an a-stem should be expected. But the
protogerm. ending nom.sg. masc. a-stem is in other runic
inscriptions preserved as -az. So, where is the -a-?
There are some three other inscriptions which show the unusual
ending -z instead of -az. In the literature they are all normally
labeled as consonant stems. Indeed this would explain the
ending -z but problems remain: first of all in all four cases
there are no counterparts either in the germ. languages nor in
other indo-european languages, so that the evidence relies
totally on the only once delivered runic inscriptions. The
interpretation as a consonant stem is therefore only given to
explain the delivered forms and cannot be verified otherwise;
the explanation seems to be circular. Secondly in the case of
gauz the derivation seems to be made quite late in language
history. Now consonant stems were in Germ. no more
productive. So a thematic stem (a-stem) should be expected.
This is also proved by the fem. derivation of the verb which
shows up as an -stem.
A closer look at the runic inscriptions reveals a more
complicated picture then is generally given. In most cases the
ending -az is still there, an ending which is continued as -r in
North Germ. (protogerm. *aaz day > run. dagaz,
northgerm. dagr). Besides there is also an ending -s which is
generally interpretated as East Germ. (protogerm. *aaz
day > run. *dags, goth. dags). At last there is an debated
ending -z; nothing seems to speak against the possibility that
this can be labeled as West Germ. (*aaz day > run.
*dagz, ohg. tag).
Most plausible interpretation: gauz [= yeller].
With the denomination the noise which the firesteel makes is
taken up.
2.b. The bog of Thorsberg
2.b.1 Sword chape

It is a sword chape of bronze
(1-2,5 mm thick) with a
length of 51 mm, an height of
47 mm and a width of 17
mm. The very well readable
runic inscription has an
height of 6-10 mm. The
object was discoverd 1860.
The deposition of the object
is dated at ca. 200 A.D. The
exact date of the inscription
cannot be given.
http://runer.natmus.dk//userfiles/im
age/ImgCache/7/rune06b_012.jpg
http://runer.natmus.dk//userfiles/image/ImgCache/7/405.JPG
I
?
()

5

10

II > ()

5

10

I o
?
wluewaz ()

5

10
II niwajemariz ()

5

10
Remark: The assumption that the Rune II,6 is not a bind-rune
but a rectified m-rune seems not very attractive.
Interpretation:
o
?
wluewaz is a masc. nom.sg. a-stem, presumably a personal
name or a surname.
The sequence ist usually interpreted as a dithematic name
consisting out of the elements o
?
wlu- and -ewaz. The last
element -ewaz servant already appeared in the name
laguewa. The first element ist connected with protogerm.
* ulu- splendour > goth. wulus, oisl. (name of a deity)
Ullr (the bright one); cp. oe. wuldor fame (< protogerm. *
ulra-).
Unclear remains the exact interpretation of the sequence o
?
w:
As a single rune can stand for its runic name, the o could be
interpreted as standing for oala herited possession
(nom.sg. neutr. n-stem) < protogerm. */alan- (besides
*/ilan-) herited possession > late goth. utal (name of the
letter in the Salzburg Ms.), ohg. uodal, osax. thil, oe. el,
el, ofris. thel, oisl. al. The the sign w would stand for
the sequence / u/ or it would be a miswriting. Although this
could be always possible, this interpretation seems to be
unlikely because in that case one would expect the name to
stand in the genitive (-ewas; cp. godagas [inscription of
Valsfjord]; asugisalas
?
[inscription of Kragehul]). The
assumption of a sequence without a syntactic connection
seems not probable.
Another explication for the missing u-rune offers the
orthographic rule that the sequence (C)VRC can be replaced
by (C) C and rendered by <(C)RC>. So the sequence <wl>
would directly stand for / ul/. But this rule is not generally
accepted and it doesnt explain the o.
It was also stated that the sequence ow is miswritten for wo,
so wolu- has to be read. The unexpected -o- in the nom.sg.
would then be an analogy from case forms that would
regurlarly show an -o- as the gen.sg. (gen.sg. protogerm.
* ula z > * olz). Of course this analogical spread ahs
left no traces otherwise.
It was also brought forth that the first sign is a non-runic sign
either marking the beginning of the inscription (without parallels)
or it would be a symbolic sign (perhaps a parallel is found on a
sword mounting found in the bog of Thorsberg).
The assumption that the first sign has not to be connected with the
rest of the inscription seems indeed probable regarding the greater
amount of space between the alledged o-rune and the w-rune
opposed to the space between the other runes. The interpretation
as a Begriffsrune (rune standing for an item) could still be
maintained assuming another date of carving (earlier or later) of
the o-rune, then of course without syntactic connection.
The sequence wl- is then most easily explained as miswriting for
wul-.
The sequence niwajemariz has most widely accepted been
divided in three words: ni, waje and mariz.
ni is the negation not < protogerm. *ni not > ohg., osax., oe.,
ofris. ni with the change of *e > *i in unaccented position
besides protogerm. *ne > oe. ne, ofris. ne, en; cp. With
secondary lengthening in one-syllabics protogerm. *n > aisl.
n.
mariz is also clearly to be connected with protogerm. *mri a-
wellknown > goth. -mereis, ohg., osax. mri, oe. mre,
oisl. mrr. To what or whom mariz refers is unclear.
In the literature it is normal to assume that because of this
inscription the adj. in its original protogerm. form has been
an i-stem protogerm. *mri-. As is true, the original
protogerm. i-stem-adj. went secondary over to the i a-stems.
But it has been overlooked that in Gothic the old i-stems still
are preserved (cp. goth. -nems < protogerm. *-nmi-
taking). So if also this adj. were an old i-stem it would have
been preserved as goth. **-mers. Goth. -mereis points
undoubtedly to an old i a-stem.
What are the consequences of this? Undoubtedly the syncope of
-a- (nom.sg. masc. mrii az) shows that the form must be of
Westgerm. origin.
This is interesting because in -ewaz the -a- is still preserved.
The difference between both forms is the weight of the
syllable which is heavy in mariz but light in -ewaz. As is
generally agreed upon the syncope of vowels took place first
in words with heavy, later in those with light syllables.
Taking this in account it could be assumed that 200 A.D. the
syncope occurred in words with heavy but not yet in those
with light syllables (other possibilities exist of course).
The new interpretation of mariz as sea presupposes the
transition of this neuter to the masculines in this early
times; also the semantic side is not convincing.
The meaning of the word waje is disputed. There are three
possible solutions:
3.conj.prs. of the weak verb I protogerm. * e/a- to
subside, to spre > oisl. vgja, a derivation of protogerm.
* i- deliberating > oisl. vgr (cp. ohg. unwgi
unimportant, mlg. unwger unfortunate), a verbal adj. to
protogerm. * ee/a- to move.
o This interpretation presupposes that <j> stands for / /,
perhaps not unparalleled.

3.conj.prs. of the reduplicated verb protogerm. * e/a- to
blow, to wave > goth. waian, (with transition to weak verbs
I) ohg. wen, ofris. wia.
dat.sg. neutr. a-stem of protogerm. * a a- bad thing > ohg.
w, a substantivation of the adj. protogerm. * a a- bad >
goth. waja- (in wajamerei* bad reputation), mhg. w- (in
wtt pain), ae. wa- (in wadd crime); cp. ohg. wwa,
wwo, oe. wwa, wa, oisl. v < protogerm. * a /an-
unfortune, danger; a derivation of protogerm. * a woe >
goth. wai, ohg., osax. w, oe. w, oisl. vei.
o Unlikely is the interpretation of wajemariz as a
dithematic word badly-wellknown because the
composition vowel should have been -a- not -e- (the
assumption of a palatalizing effect of is unparalleled).
Most plausible interpretation: Inherited possession.
?

Wuluewaz [= shining-servant]. Not may the wellknown
harm/The not because of bad things wellknown.
3. THE EARLY RUNIC CLASPS
3.a. The clasp of Grdlsa
The silver clasp has a length of 42 mm, the length of the clip is
47 mm. The clasp was found in the grave of a woman of
higher social rank. The object is dated around 200 A.D. (a
new dating places the object in a time zone from 200 till 350
A.D.).

http://runer.n
atmus.dk//us
erfiles/image
/ImgCache/6
18/rune02b_
243.jpg
()

5

ekunwod ()

5

Remarks: The last rune has no counterparts and is not
identical with other runic signs, so that a transliteration is
difficult. Because of general considerations and on formal
grounds the interpretation as z-rune lies on the hand.
Interpretation:
ekunwodz has to be divided in two words. The segmentation
lies between -k and -u.
ek is the personal pronoun of the 1
st
person sg. < protogerm. *ek
I > goth. ik (if not < protogerm. *ik), oisl. ek; besides (with
*e > *i in unaccented position or analogical to the acc.sg.
protogerm. *mek
e
> late protogerm. *mik
i
) protogerm. *ik >
run. ik, goth. ik (if not < protogerm *ek), ohg. ih, osax. ik, oe.
ic, ofris. ik; besides protogerm. *eka
n
> run. -eka, run.-dan.,
onorw., oswed. iak and (again with *e > *i) protogerm. *ika
n

> run. -ika; cp. also ohg. ihha I myself (< protogerm. *ik).
unwodz nom.sg.masc. adj. a-stem, dithematic word, perhaps a
byname < protogerm. *un a- not raging, consisting out
of the elements un- and -wodz:
un- < protogerm. *un- un-, not (negation particle) > goth. un-,
ohg., osax., oe., ofris. un-, oisl. -, - (cp. in personal names
perhaps germ.-lat. Untancus
?
).
-wodz < protogerm. * a- angry, raging, furious > goth.
wos (-d-), oe. wd, oisl. r; cp. the derivation ohg. wuoten
(weak verb I) to rage, to be angry; cp. also run. wodu-, oisl.
(name of a deity) r < protogerm. * u- anger, rage.
The classification as an a-stem can be defended, because the
compositions with un- regularly are in the same stemclass as
the uncompounded words.
Because of the lacking -a- often a consonant-stem is
supposed. But adj. are in Germanic not consonantal. So
this assumption is unprobable.
Syntactic remarks: in the order personal pronoun adj. has the
adj. the function of an apposition, which qualifies the
personal pronoun.
It is the oldest germ. record of a selfpredication.
Speculative remain further interpretations of unwodz, so as a
functionname or a pseudonym of the carver of the
inscription. A mere personal name can at least not be
excluded.
Most plausible interpretation: I, unwodz [= not
angry/raging/furious].
3.b. The clasp of Himlingje II

age/ImgCache/619/tegning.JPG
The silver rosette clasp has a
length of 107 mm Lnge, a width
of 116 mm and a height of 50 mm.
It was discovered 1949 and is
dated 200 A.D. The runes with a
total length of 30 mm were
inscribed on the catch-plate,
which is not completely
preserved. The clasp was found in
the grave of a woman of higher
social rank.
()

5

iduhudaz ()

5

Remarks: Because the clasp is damaged parts of the runes 1-3
are lost. It is unknown whether before rune 1 there were
more runes inscribed.
Interpretation:
nom.sg. masc. a-stem < protogerm. * iuunda-, a dithematic
word, arguably not a personal name, containing the elements
* iu- and *-unda-.
Protogerm. * iu- wood > ohg. witu, osax. widu, wido, oe.
widu, wudu, wiodu, aisl. vir.
Protogerm. *unda- dog > goth. hunds*, ohg. hunt, osax., oe.
hund, ofris. hond, oisl. hundr.
For the composition cp. mhd. holzhund, walthund wood-dog.
So the etymology is not problematic. Disputed is however what
is exactly meant bei wood-dog. It seems clear that it is a
designation for wolf. But the ususal analysis as name of the
rune-carver or as a personal name is not probable because the
element protogerm. *unda- doesnt show up in personal
names otherwise. Perhaps it is to be understood as a
metaphor in the function of a personal designation standing
for a so called berserkr or ulfheinn, so for a wolf-warrior.
Most plausible interpretation: Widuhundaz [wood-dog].
3.c. The clasp of Lundegrde/Nvling
This is a silver rosette clasp with a golden layer on the rosettes.
The inscription was carved on the catch-plate, which is ca. 20
x 40 mm.). The clasp was found in 1963 in the grave of a
woman of higher social rank. The inscription was discovered
later during the conservation of the corroded material. The
grave find is dated at 200 A.D.
http://runer.natmus.dk//userfiles/image/ImgCache/620/246.JPG
()

5

10 15
bidawarijaztalgidai
5

10

15

Remarks: Remarkably the inscription was carved on the outside
of the catch-plate. Also notable is the fact that the Rune 9 has
the total lineheight.
Interpretation:
bidawarijaztalgidai has to be divided in two words. The
segmentation lies between z- and -t.
bidawarijaz is a nom.sg. masc. a-stem, personal name <
protogerm. *ia ari a-, as can be seen easily with more
possibilities in the first element:
protogerm. *a- waiter
?
, a derivation with the deverbative
nouns formative suffix protogerm. *-a- of protogerm. *e
e/a- to wait > goth. beidan, ohg. btan, osax. bdan, oe.
bdan, ofris. bidia, oisl. ba.
protogerm. *i a- applicant
?
, also a derivation with the
deverbative nouns formative suffix protogerm. *-a- of
protogerm. *i e/a- to beg, to pray > goth. bidjan, ohg.
bittan, osax. biddian, oe. biddan, ofris. bidda, aisl. bija.
protogerm. *enda- bond
?
, also a derivation with the
deverbative nouns formative suffix protogerm. *-a- of
protogerm. *ende/a- to bind > goth. bindan, ohg. bintan,
osax., oe. bindan, ofris., oisl. binda.
improbable: interpretation of bida- as protogerm. *a-
neutr. a-stem pact, matrimony and connection with lat.
foedus pact; problem: no other derivation of protogerm.
*e/a- has the meaning pact.
The second element -warijaz < urgerm. * ar a- defender,
protecter > lat.-germ. -varii, a derivation with the
deverbative nouns formative suffix protogerm. *- a- of
protogerm. * ar e/a- to defend, to protect > goth. warjan,
ohg. wer(r)en, osax., oe. werian, ofris. wera, aisl. verja.
Problems of interpretation: when bdawarijaz is read the
meaning of the name is not clear (a defender/protector of the
waiter makes no sense at all). Wenn instead bidawarijaz is
read the meaning makes sense (defender/protector of the
applicant) but the composition vowel is not explicable; to be
expected were -ja- (cp. aljamarkiz [stone-plate of Krstad);
of course the composition vowel can change in Germanic.
The second word talgidai must of course (cp. the sequence
niijo tawide) be a verb.
3.sg.ind.pret. of the verb protogerm. *tali e/a- weak verb I to
carve, to cut, to incise > oisl. telgja.
Because the ending -dai is thought to be problematic, some
assumptions have been put forward to get rid of it:
talgidai should be split in talgida i carved in; problematic is
that the juncture *tali e/a- en is without parallels and the
form protogerm. *en should have been written as in in this
early time.
talgidai should be read as talgida |; the last sign would be a
division mark; problem: other division marks that look like i-
runes cannot be adduced.
Quite absurd is the interpretation of the sequence talgidai as
talgida ai the carver alas!; talgida would be a masc. n-
stem the carver, a derivation with the suffix protogerm. *-
(a/i)/an- of the verb protogerm. *tali e/a- to carve;
problem: this suffix builds only abstracta.
Most plausible interpretation: Bidawarijaz carved most
probable is a runeic carver-inscription.
3.d. The clasp of Vrlse
This is a silver rosette clasp ornamented with a gold-plated
silver steel sheet and 7 rosettes. with a golden layer on the
rosettes. The clasp measures 90 x 96 x 40 mm. The runic
inscription with an height of ca. 7 mm. The clasp was found
in 1944 in the grave of a woman of higher social rank. The
grave find is dated at 200 A.D.
http://runer.natmus.dk//userfiles/image/ImgCache/616/V%C3%
A6rl%C3%B8se.JPG
http://runer.natmus.dk//userfiles
/image/ImgCache/616/rune06b
_224.jpg
The runic inscriptions is carved
in the free area of the
ornamented needle holder. These
ornaments consist out lines and a
swastika. The runes themselves
are evidently carved later the the
ornaments because they are cut
in them.
()

5
alugod
5

Remarks: The o-rune takes only the 2/3 lineheight. The d-rune
is very compact because of the following swastika. The o-rune
overlaps the lower part of the g-rune.
Interpretation:
alugod seems in light of the other inscriptions on clasps to be
interpreted as a dithematic personal name, that has to be
divided in the elements alu- and -god.
The element alu- is somewhat complicated, but is usually
connected to protogerm. *alu
inebriant, beer > mhg. al- (in

alscaf drinking vessel), osax. alu- (in alufat beer vessel),
oe. ealu, oisl. l.
The element -god is to be connected to protogerm. *a-
good > goth. gos, ohg. guot, osax., oe., ofris. gd, oisl.
gr.
The last element is a bit problematic, because protogerm.
*a- good seems otherwise not to appear as the second
element in Germ. personal names.
Much more problematic: there is no ending. How to explain?
Interpretation of alugod as voc.sg.masc.; formally: this
interpretation explains the non-ending. But: when vocative, a
fem. vocative would be expected (the clasp was found in a
womans grave; assumption that the clasp was formerly worn
by a man as purely circular) and a fem. vocative would not
have been endingless. On the other hand: vocative forms in
runic inscriptions cannot be secured (the alledged parallel
forms [alawin (thrice) and alawid on the bracteate B of
Skodborghus] shouldnt be taken into account because of the
difference of 275 years; of course these forms can be
explained as westgerm. nom.sg. with total fallout of the
former ending).
alugod is a westgerm. nom.sg. ending with early fallout of
the total ending (also to be compared with alawin and
alawid); also against this speaks the difference in time: a so
early complete fallout of the ending has no parallels (indeed
it seems to be that around 200 A.D. the westgerm. ending of
the masc. a-stems was -z).
alugod is no personal name but a composite word consisting
of alu und god; alu would be the word for beer or more
general feast; god would be a neutr. consonant stem with
the meaning offering and to be connected with ne. god;
problem: this word was not a consonant stem but a neutr. a-
stem, so the endingless form is not explained. More severe is
perhaps the fact, that a personal name has to be expexted,
otherwise this inscription would fall out of all other
inscriptions on clasps
So how can the endingless form be explained? It is easy to be
seen that the first three signs were written with considerable
space. Against this the last three signs stand very narrowly,
even overlapping. Also the next to the last rune is smaller
then the rest of the runes, so that the upper part of the o-rune
is adapted to fit in the croop of the g-rune. It is quite obvious
that the reason for this is shortage of space. An o-rune as big
as the other runes would had as consequence that it should
have been written further to the right, but there would have
been no space to do that. Also the last rune is written very
condensed and very near to the swastika.

These observations allow a plausible interpretation for the lost
ending: the runecarver didnt get the inscription fit in the
area. Because no place was there, he just omitted the ending.
To omit the ending was of course less dramatic then leaving
out the d-rune which would have been necessary writen the
o-rune as big as the other runes. So only the diminshment of
the o-rune made it possible to write the d-rune. Under this
assumption it lies on the hand that it is more probable that
the ending consisted only out of one additional sign. So the
most probable ending is either -z, -a, or -o (*alugodz,
*alugoda, *alugodo) not -az (*alugodaz).
Most plausible interpretation: Alugod [beer-good].

3.e. The clasp of Skovgrde/Udby
This is a silver rosette clasp ornamented with a gold-plate silver
steel sheet. The clasp measures 111 x 101 x 48 mm. The
runic inscription with an height of ca. 10 mm. The clasp was
found in 1988 in the grave of a woman of higher social rank;
the inscription was only discovered 1990. The grave find is
dated at 200 A.D. The runic inscription is unique because it
shows two different writing directions.
( : )

5 10
lamo:talgida

5 10

Remarks: This is the only archaic inscription that shows a
change in the writing direction (quasi-boustrophedon). The
best explanation for this is to be found in the fact that it was
copied from a template that was written in boustrophedon.
Interpretation:
The inscription consists of two words divided by a word
marker.
lamo nom.sg. m.asc/fem. n-stem, surely a personal name (but
whether shortened or nickname cannot be decided formally)
< protogerm. *laman-/n- the lame one > osax. lamo, oe.
lama, ofris. loma, oisl. lami, a derivation with the persons
designating suffix protogerm. *-n- of protogerm. *lama-
lame > ohg., osax. lam, ofris. lom.
talgida is the 3.sg.ind.pret. of the verb protogerm. *tali e/a-
weak verb I to carve, to cut, to incise > oisl. telgja.
DIGRESSION: THE ENDINGS OF THE 3.SG.IND.PRET.

As seen in the 3.sg.ind.prt. of the weak verbs there show up 3
endings:
run. -dai: (rosette clasp of Lundegrde/Nvling; ca. 200) talgidai
(bidawarijaz talgidai Bidawarijaz carved).
run. -de: (shield mount 2 of Illerup dal; ca. 200;) tawide (niijo
tawide Niijo made), (wooden box of Garblle; ca. 400) tawide
(hagiradaz : tawide : Hagiradaz made), (bracteate 1 of Tjurk;
ca. 500) wurte (wurte runoz an walhakurne it appeared the
runes on the welsh-grain [= gold] ).
run. -da: (rosette clasp of Skovgrde/Udby; ca. 200) talgida (lamo
: talgida Lamo carved).
As a variance between -ai and -e appears in the dat.sg. of the a-
stems too, this received most attention.
For the coexsistence side by side many proposals have been
made to explain it:
Graphical mistake: an e-rune in the template (form: ) was
miscarved as ai ( ): the misreading is so elementary that
there is absolutely no need to start speculating on the
possibility of an archaic verbal form. Although as
miscarving can at no times be excluded, a repeated
miscarving with the same result is improbable.
Also graphically: assumption that run. -ai, -e and -a are
graphical attempts to render one phoneme [:]. Against
this speaks the fact that it would be the only case that one
phoneme is rendered by more runic signs (there are no other
examples).
The reading talgidai has to be altered in talgida |, so with a
word final division mark. There would therefore be only two
endings. But a word division mark in form of an i-Rune
would be singular.
All three endings would be correct: -e would be the
continuant of protogerm. *-
/
, -a an analogious form after
the 2.sg. *-z and -ai in the end the continuant of protogerm.
*-a . But on the one hand side the temporal concomittance is
not regarded, on the other hand side it would be a
spontaneous analogy that would have left no traces in
language history.
The most widespread explanation: in the variance between
run. -ai : -e (in the verb and noun) an historical change
(monophthongization) can be seen. Protogerm. Unaccented
*-a became run. e. run. -da would point to protogerm. *-d
d
,
run. -de < run. -dai to preprotogerm. *-ti.
Problems: Chronological problem 1: the concommitance of
run. -ai und -e in the nearly simultaneous inscriptions on the
rosette clasp of Lundegrde/Nvling and on the shield mount
2 of Illerup dal is not explained (no proponent of the
monophthongization-theory dwelled on it). Chronological
problem 2: in the dative in an inscription ca. 400 the form
hahai is found (with -ai); this is not to be expected. This late
-ai is regularly explained as a reversal writing, so it comes
from the time before the monophthongization; but: to keep
an old writing as a conservative writing it is necessary that
much is written (better rewritten) or the grammarians occupy
themselves with writing rules. Both is improbable for these
early times.
Now also the early transmission of the younger -e speaks
against this assumption. Because not only the late -ai but also
the early -ai in talgidai should be the result (simultanious
with -e in tawide) of an archaic writing. So the
monophthongization must be dated before 200 and therefort
before runic writing. It is not possible that a writing before
writing was introduced was preserved. Also the assumption
of an accidental transmission of run. -ai und -e during the
monophthongization-process seems to be far-fethced.
As it seems that no explanation catches the truth, it must be
asked if anything else is possible.
As is clear the simultanious forms -dai and -de cannot be
explained by a phonological change of *-a to *-. So it is not
possible that both forms continue preprotogerm. *-to .
Therefore run. -da cannot continue protogerm. *-
because
the the ending run. -de would be unexplained. Because -da
cannot continue protogerm. *-
and an analogy is not

probable, another explication for -da must be found.
(herewith the strongest argument that run. -a in the masc. n-
stems continues protogerm. *-n < protoie. *-n falls apart).
So lets again have a closer look at talgida.
As is clear the simultanious forms -dai and -de cannot be
explained by a phonological change of *-a to *-. So it is not
possible that both forms continue preprotogerm. *-to .
Therefore run. -da cannot continue protogerm. *-
because
the the ending run. -de would be unexplained. Because -da
cannot continue protogerm. *-
and an analogy is not

probable, another explication for -da must be found.
(herewith the strongest argument that run. -a in the masc. n-
stems continues protogerm. *-n < protoie. *-n falls apart).
So lets again have a closer look at talgida.
Because the runic carver misjudged the space it seems possible
the he left out a rune because of space problems. Because a
further runic sign it must have been an i-rune could not
have been written by the carver. As is the case on larger
objects the addition of the rune above the line was her not
possible. Between the a-rune and the division mark there was
no space to inscribe the i-rune. To write the i-rune after the
division mark would not have changed talgida in talgidai but
rather lamo in lamoi. The change of the division mark into an
i-rune would have made the whole inscription totally
illegible. So it is conceivable that the runic carver chose the
easiest way out of the problem and just didnt write the i-
rune.
So it can be assumed that the form talgida doesnt show a
separate third ending of the 3.sg.ind.pret. and is also no reals
miswriting. It is just the omittance of a rune due to space
problems. talgida is therefore to be understood as *talgida(i).
For the early runic inscriptions again only two endings exist: -
dai and -de. That the coincident transmission shows them to
be two separate endings seems clear. And indeed: the
Germanic parent language gives us also only two
endings:protogerm. *-
/
and protogerm. *-/a . Die
erstere, urgerm. *-
/
, liegt dabei in run. -de, die letztere,
urgerm. *-/a dagegen in run. -dai.
Because the old short unaccented diphthong -ai is still preserved
in the earliest runic inscriptions, it is clear the the
monophthongization to - can only have ocurred afterwards,
after 200 A.D.

4. THE MAGIC-POETICAL LANCENAMES
4.a. The lancehead of vre Stabu
Die lanceheand of the type Vennolum has a length of ca. 28 mm and
shows because of burning and later corrosion strong damages in
form of holes. The runic signs have a maximal height of 10 mm
and are carved by means of fine parallel strokes. They are in a
frame consisting out of a dotted line which becomes smaller. The
graves were found already 1890 but the runic inscription was
discovered only 1900. The lancehead type points to a dating
between 200-250. The dating of the inscription cannot be secured,
although the type of carving seems to suggest that the inscription
was not carved immediatly before the lancehead was laid down.

http://www.arild-hauge.com/arild-hauge/no-rune-oevere-stabu1.jpg
http://www.arild-hauge.com/arild-hauge/no-rune-stabu.jpg
()

5
raunijaz

5

Remarks: The last rune is so stronly damaged that only a staff
can be seen. Of course the rendering with z is without doubt.
Interpretation:
nom.sg. masc. a-stem < protogerm. *ra ni a- tryer, tester >
oisl. reynir tester, a derivation with the verbal nouns
formative suffix protogerm. *- a- of the verb protogerm. *ra
ni e/a- to try, to test > oisl. reyna. A different derivation
shows up in oisl. raun try, test, experience, ofris. rn
?

investigation < protogerm. *ra n-.
Most plausible interpretation: tryer.
4.b. The lancehead of Dahmsdorf
This is an iron lanchead of the type Lynghjgrd with a length
of 169 mm and a bladewidth of 43 mm. All runes and
symbolic signs were brought on the lance with silver wire;
because of influence of fire this melted partly out. The side
with the inscription shows a circle, a half-moon symbol and a
double-fork (Tamga), also interpreted as a thunderbolt. The
side without runes shows symbols too. Perhaps some of the
symbols could point to the Black Sea area.

The lancehead was found 1865 and was formerly in the
museum of Mncheberg but is lost since the end of world
war II. Lances of this type were in long use but this on is
dated on ground of other findings to 250-320.

http://www.gotica.de/runica.html
()

5
ranja

5

Interpretation:
nom.sg. m. an-stem < protogerm. *ranni an- runner, not
continued in a germ. language. It is a derivation with the
deverbative nouns formative suffix protogerm. *-( )an- of the
verb protogerm. *ranni e/a- to make walk > goth. -rannjan,
ohg. rennen, os. rennian, oe. -rennan, ofris. renna, rinna,
oisl. renna, itself a causative verb derivated from protogerm.
*renne/a- walk > goth. rinnan, ohd., osax., oe. rinnan,
ofris. renna, rinna, oisl. rinna.
Other interpretations seem to be improbable:
It would be an imitation of the Roman fabrication name
RANVICI (found on a sword from Einang).
It would be faulty copy of the inscription raunijaz from vre
Stabu.
Problem in both cases: the probability that an imitation or faulty
copy would render a word and even a word making sense is
very small.
Most plausible interpretation: runner.
4.c. The lancehead of Kowel
The lancehead has a length of 155 mm and ist broadest part is
30 mm long. It has on both sides ornaments (symbolic signs)
that are like the runic inscription brought to the lancehead
with silver wire. The runes have an height of 6 10,5 mm
and the inscription is 35 mm long. The lancehead was found
1858 but first read 1875. The lancehead was brought to
Berlin during world war II but disappeared from there on a
transport to Oberfranken.
The dating of the inscription is difficult because there is no
finding context. The symbolig signs point to end of the 3
rd
or
beginning of the 4
th
century.
http://www.arild-hauge.com/ru-e-rusland.htm
()

5

tilarid
?
s

5

Remarks: Krause discovered on the non-runic side of the
lancehead an o-rune which he interpreted as inherited
possession. Problematic is the next to last rune which has no
counterparts.
Interpretation:
nom.sg. masc. a-stem, dithematic word consisting out of the
elements tila- and -rids; < protogerm. *tilara-.
The element tila- can be connected morefold:
to protogerm. *tilan to, at > oe., ofris., oisl. til (in fact this
is an acc.sg. of the last word).
to protogerm. *tila- apt > goth. (only adv.) -tilaba, oe.,
ofris. til, oisl. tilr.
to protogerm. *tila- goal > got. til, ahd. zil (cp. oisl. -tili <
protogerm. *til a-).
-rids < protogerm. *ra- equestrian, rider > run. -ridaz and -
ride, a derivation with the verbal nouns formative suffix
protogerm. *-a- of protogerm. *re/a- to ride > ohg. rtan,
os. -rdan, oe. rdan, ofris. rda, oisl. ra.
The also proposed connection of the element -rids with the
wordgroup protogerm. *re/a- to guess must be dismissed
because of phonological reasons.
The comparison with oe. Personal name Tilred has to be given
up because Tilred is a late composition.
Because the reading of the next to last rune is not secured there
habe been other readings:
ikharis standing for ik harings I, Haring.
eharis standing for ek harings with an abbreviated
personal pronoun (this has parallels, cp. eerilaz, generally
interpreted as ek erilaz).
reading as non-runic but an illyrian personal name Tilarios.
General problem: why tilarids, why no tilaris?
Most plausible interpretation: goal-router.
DIGRESSION: THE RUNIC ROW
The runic row is only two times transmitted complete: on the
one hand side on the stone plate of Kylver (after 375,
probably 5
th
century), on the other hand side on two model-
identical (same stamp) swedish bracteats (both ca. 500-550)
of Vadstena-C and Raum Mariedam-C:
Kylver fuarkgwhnijpzstbemldo
Vadstena-C fuarkgw:hnijbzs:tbemlo:
Raum Mariedam-C fuarkgw:hnijbzs:tbemlo:


http://en.wikipedia.org/wiki/File:Kylverstenen_2.jpg

http://www.sitecenter.dk/schleu.dk/nss-folder/aidalatar/vadstena.jpg
The runic row is more often transmitted incomplete, for
example:
bracteate of Grumpan-C (ca. 500-550):
fuarkgw..hnijp**.tbemlod
pilar of Breza (ca. 500-600): fuarkgwhnijpzsteml[
clasp ofCharnay (ca. 550-600): fuarkgwhnijpzstbem
All other incomplete runic rows are much shorter. The only
inscription of importance here is that on the clasp of
Aquincum (ca. 500): fuarkgw.
http://www.arild-hauge.com/arild-hauge/196_Vg207_Savare_SHM.jpg
The runic row is in two separate inscriptions (on the bracteates
of Vadstena-C and of Raum Mariedam-C [same stamp] and
on the bracteate of Grumpan-C) divided in three groups
containing 8 runes each. The first group of 8 runes is also
found on the clast of Aquincum.
These groups are named as ttir, (sg.) tt Geschlecht, family.
This designation is very recent, it is transmitted from the
17th centurey in icelandic manuscripts. The three groups are
called after the first rune in each group Freys tt, Hagals tt
and Tys tt.
Between the complete runic rows of the stone-plate of Kylver
and the bracteates of Raum Vadstena and Raum Mariedam
there is a a difference in the sequence of the runes. The
runes13/14 and 23/24 appear on the stone-plate of Kylver als
p/ and d/o, on the bracteates of Raum Vadstena and Raum
Mariedam against this as /p and o/d. With this last sequence
all other older runic rows are in accord (when they show
these runes). The sequence d/o as on the stone-plate of
Kylver is confirmed by the
old english fuorc on the Themse-knife (9
th
century). Though
the stone-plate of Kylver stands alone in the early
transmission it must not be overlooked that this is the oldest
runic row.
In far the most standard introductions there is found a
compromise between both runic rows: the runes 13/14 are
given as /p the runes 23/24 in contrast as d/o. This should be
given up, because it is a pure fictional runic row.
Because the transmission shows no unity it can be asked which
is the older sequence of the fuark. Combining the
transmitted runic rows one could come to this:
fuarkgw // hnij/pzs // tbemlo/d
This gives two possibilities:
CV CV (stone-plate of Kylver: p do)

VC VC (bracteates of Vadstena/Motola: p od)

Because the insecurity concern the middle and the end of the
runic row it could be possible that the division in three ttir
is not the oldest one, but that a division in two parts has to be
assumed:
fuarkgwhnij
/pzstbemlngo/d

So there seems to be a symmetry between either the beginnings
or the endings of both parts:
beginning end
first half: fu ij
second half: Kylver: p do
andere: p od
Both runic rows are therefore seen from a symmetrical
viewpoint of the same value; therefore there seems to be no
argument to prefer one runic row above the other.

DIGRESSION: THE BEGINNING OF RUNIC
WRITING?
There does exist a very early inscribed clasp, the clasp of
Meldorf (dated at 25 A.D.). The characters are inscribed by
means of fine parallel strokes/dots.
http://upload.wikimedia.org/wikipedia/commons/f/fa/Meldorf_inscription_d
rawing.jpg
Problem: what script are these characters? It is posed that they
are runes, protorunes or even just Latin script.
Possible readings are: hiri/irih//hiwi/iwih//idin/nidi//iih/hii
(// [?]).
Of course not all readings make sense. There seem to be only to
possible readings:
hwi: dat./acc.sg. fem. i-stem < protogerm. * i-. Problem:
in the Germ. languages no i-stem is transmitted but only an
a-stem protogerm. * a- house > goth. heiwa-frauja
house-lord, oe. hw-cund domestic.
Besides there is a n-stem protogerm. * /an- > ohg. ha
spouse, ho spouse; a collective protogerm. Substantive
derived from an adj. * -iska- (with the adj. forming suffix
protogerm. *-iska-) > osax., ohg. hiwiski, oe. hwisc, oisl.
hski family < protogerm. * -isk- a-); a derivation with
the suffix protogerm. *-ia- in run. (stone of rstad, ca. 550)
hiwigaz personal name < protogerm. * -ia-. One would
have to accept a fem. i-stem protogerm. * iz spouse
(perhaps here as personal name). Formally a dat.sg. should
be taken into account, so for the spouse/for Hiwiz. But: in
accord to the relation * an- : * n- next to * a- rather a
fem. * - and not * i- shoudl be expected.
idin: dat.sg. fem. n-stem of protogerm. *in-, a personal
name. This is quite an improbable interpretation, because the
dat.sg. fem. n-stem should have looked like **idn <
protogerm. *ini. Phonologically more plausible is the
interpretation as an dat.sg. masc. n-stem of the masc.
counterpart *ian- because this would have looked like
protogerm. *ieni that would have become idin.
Because no interpretation seems to be very convincing it was
assumed that the characters are no real letters but is rather an
imitation of script.
5. LATER BOG FINDS
5.a. The sword chape of Vimose
This sword chape consists of two bronze plates and shows a
nearly circular form. It measures ca. 32 x 27 mm and shows
two rivet holes. Which of the two sides were on the front is
unclear. The sword chape was discovered 1901; its dating is
between 200-250.
http://runer.natmus.dk/
/userfiles/image/ImgC
ache/201/rune06b_15
5.jpg
ache/201/rune06b_15
6.jpg
http://runer.natmus.dk//userfiles/image/ImgCache/201/tegning.JPG
A 1 ()

5

2 ()
B ()

5

A 1. mariha
2. iala
B makija

Remarks: The reading of rune A 1,5 as a d-rune is not possible.
Interpretation:
The interpretation is difficult because no word division marks
show up; so the exact segmentation of the words remains
unclear. Proposed segmentation: marihai ala(-)makija.
marihai could be dat.sg. m. a-stem, perhaps a personal name or
a surname < protogerm. *mri a- the famous one, a
substantivation of the adj. protogerm. *mri a- famous >
run. -mariz ( sword chape of Thorsberg), got. -mereis* (in
wailamereis* laudable), ohg., os. mri, oe. mre, oisl.
mrr.
Problem: why was the protogerm. * written with run. h? Under
the assumption that run. h stands for the sound [] then it
could be assumed that because the run. j standing for the
sound [j] differs only in voicing the [j] of the suffix could
be written with run. h; that the carver had probably problems
with the writing of medial [j] seems also to show the form
makija because here the <j> is but later inserted.
Because the ending -ai finds itself only in the paradigm of the
substantives (in the paradigm of the adjectives there is in this
case a pronominal ending) it must be assumed that the form
marihai ist to be interpreted as a substantivized adjective.
Of course other interpretations exist:
division of the sequence marihai in two words either as mari
hai or as mari ha. mari hai would stand for mari aih (s)he
possesses the famous (one) (so with miswriting of the rune
h) and mari ha would stand for mari habe (s)he may have
the famous (one). Although in principle this could be
possible it must be dismissed because the adjective is an old
a-stem; so it cannot be written only mari.
interpretation as extension with the suffix -ha < protogerm.
*-a- is because of this reason equally imporbable (also
appears the suffix more regularly as protogerm. *-a-).
the reading is maridai and interpretation as a 3.sg.ind.pret. is
not possible because no d-rune can be read.
Other possible solution: segmentation in mari hai and reading as
mari aih when mari is connected with protogerm. nom.sg.
neutr. i-St. < protogerm. *mari- sea, lake > goth. mari-,
ohg., osax. meri, ofris. mere, oisl. marr.
The sequence alamakija is controversial because it is not secure
whether it is one word or has to be divided in two words (ala
makija).
When two words are assumed:
alla is a nom.sg. masc. n-stem, personal name (shortened out of
a dithematic name) < protogerm. *allan- the complete one
often ocurring as personal name, cp. goth. Alla, ohg. Allo, oe.
Alla, oisl. Alli, a derivation with the individualizing suffix
protogerm. *-n- of protogerm. *alla- < *alna- all, whole >
goth. alls, ohg. al(l), osax. all, oe. eall, ofris. al(l), oisl. allr,
itself a derivation with the verbal adjectives formative suffix
protogerm. *-na- of protogerm. *ale/a- to nourish > goth.,
oe. alan, oisl. ala.
makija is an acc.sg. masc. a-stem < protogerm. *mki a-
sword > goth. mekeis*/meki*, krimgoth. mycha, osax. mki,
oe. mce, oisl. mkir.
When one word is assumed:
alamakija is an acc.sg. masc. a-stem < protogerm. *alamki a-
all-sword, a composition out of protogerm. *ala- in
strengthening function > lat.-germ. Ala- (in EN Alamanni),
goth., ohg., osax. ala-, oe., ofris., oisl. al- and protogerm.
*mki a-; perhaps a collective function has to be assumed.
Not verifiable is the interpretation of makija as a nom./acc.sg.
neutr. a-stem, though it cannot be excluded totally on the
basis of goth. (acc.sg.) meki. Under this assumption ala could
be assumed to be a nom./acc.sg.neutr. adj. a-stem protogerm.
*allan all, total and be connected with makija or be the first
element of a compound with strengthening function.
Syntactically this inscription though long cannot be used.
Problematic remains the fact that (perhaps) no verbform appears
in the inscription. When no verb is there the interpretation of
the sequence alamakija as two words must be preferred
because otherwise the inscription would not have a
nominative too.
Most plausible interpretation(s): Alla <gives, offers?> the
famous one the sword // the famous one <> the all-sword //
the lake has the all-sword.
5.b. The buckle of Vimose
It is a bronze buckle with a length of 54 mm, a width of 50 mm
and a height of 8 mm. The inscription is carved on the side
not seen when wearing the sword. The plate that bears the
inscription measures 40 x 29 mm. The object has rivets and
the inscription must be carved after the final assembly
because the runes are incised in them. The object was
discovered already 1851 but the inscription not before 1868.
Typologically the object is dated between 200-250. Because
it is a mobile object the place and time of the inscription
cannot be secured.
age/ImgCache/205/rune06b_170.jp
g
http://runer.natmus.dk//userfiles/ima
ge/ImgCache/205/rune06b_172.jpg
http://runer.natmus.dk//userfiles/image/ImgCache/205/tegning.JPG
1 ()

5

2 ()

5

10

1. aadagasu ()

5

2. laasauwija ()

5

10

Remarks: The reading of rune 1,8 as a t-rune is not possible.
Interpretation:
The interpretation is difficult because on the one hand side no
word division marks appear and on the other hand side there
is an unexpected vowel-doubling; so the exact segmentation
of the words remains unclear. An indication for a
segmentation is given by the doubled vocalic runes. Because
there is no inscription at all in which vowel length is
rendered by doubled runic signs the only plausible
conclusion is that between them there is a word boundary, so
one word ends in a vowel and the other one starts with a
vowel. This is generally assumed for line 2 so this
interpretation must also be taken over for line 1.
Proposed segmentation: a ada g asula asau wija. Of course a
and g are no words so it must be assumed that they stand
for their runic names.
a stands for ist runic name, without secure identification of the
case; protogerm. masc. u-stem *ansu- god > run. asu-, late
goth. aza (< *anzaz < got. *ansus), ohg. ans(i)- in personal
names, osax. s-, s- in personal names, oe. s, oisl. ss; cp.
Also lat.-goth. (acc.pl.) ansis (i-stem).
The impossible reading of rune 1,8 as a t-rune lead to an
interpretation aadagast of the complete first line which was
interpreted as a miscarving for a personal name andagast.
g stands for ist runic name, without secure identification of the
case; protogerm. *e- gift > goth. giba, ohg. geba, osax.
gea, oe. giefu, ofris. jeve, oisl. gjf, a derivation with the
abstracts of verbs formative suffix protogerm. *-- of
protogerm. *ee/a- to give > goth. giban, ohg., osax.
geban, oe. giefan, ofris. ieva, oisl. gefa.
When the interpretation of these two runes is accepted (not
uncontradicted) the element in between is most likely
perceived as a conjunction joining the two terms:
a<n>da conj. < protogerm. *anda and > oe. and, ond, ofris.
an(d); cp. protogerm. *and/i > ohg. anti, enti, osax. endi,
oisl. en; protogerm. *und > ohg. unti; protogerm. *und/a >
ohg. unta.
The sequence adag has also been interpreted as an acc.sg. of a
personal name (or even as a nom.sg. without explication of
the missing ending), a derivation with the suffix protogerm.
*-aa- of protogerm. *andan- breath(ing) > ohg. anto,
osax. ando, oe. anda, oisl. andi. Problematic remains that
this element does not appear in personal names (the adduced
name lat.-east goth. [gen.sg.] Andagis is surely a dithematic
name, namele Anda-gis.
a<n>sula is a nom.sg. masc. n-stem personal name <
protogerm. *ansulan- the one belonging to god/the little
god > lat.-germ. Ansila, oisl. sli, a derivation with the
personal names formative suffix protogerm. *-n- of *ansula-,
a affilation or deminuative formation with the suffix
protogerm. *-la- of protogerm. *ansu- god.
Less probable is the interpretation of asula as buckle with
wich lat. ansula small handle is compared. The root is
preserved in Germ. in oisl. s eyelet, mlg. se sling <
protogerm. *ansi - but a la-derivation cannot be secured and
the meaning buckle seems to be extrapolated from the
object itself.
a<n>sau: dat.sg. masc. u-stem < protogerm. *ansu- god.
Because the double signs do not stand for long vowels, the laas-
cannot be interpreted as ls-.
Also not probable is the interpretation of -laas as a miswriting
for -laus, so asulaas stood for ansulaus godless. On the one
hand the reason for the miswriting cannot be seen, on the
other hand the semantic motivation for the epithet would be
unclear (no argument against this interpretation is the form -
laus itself, because assuming an East Germ. inscription the
expected -ss would have been reduced to -s).
wija could be a 1(3.
?
).sg.ind.pres. weak verb III < protogerm.
* /a /- e/a- to consecrate, to dedicate > run. wiju and
perhaps wihgu
?
, goth. weihan, ohg. when, osax. whian,
ofris. w(g)a, oisl. vgja, a derivation of protogerm. * a-
holy, sacred > goth. weihs, ohg. wh, osax. wh-; cp. The
substantivation osax. wh temple, oe. wh image of a
deity, oisl. v sacred place, court place (< * a-).
Problem: where is the expected -- that should appear as -h-?
This could have a parallel in the form wiju. This loss could
have been produced by the phonetic status of both sounds.
When it is assumed that protogerm. * could be written with h
and j it could be thought that also the combination * could
have been written with -jj- which according to the
orthographic rule was simplified to -j-.
Not possible is the interpretation of the sequence auwija as a
form with already occurred Westgerm. gemination because
this has to be dated later.
The language of the inscription remains unclear; the ending -a
when interpreted correctly seems to point to an East
Germanic language.
Most plausible interpretation: God and gift. (I) the one
belonging to god/the little god dedicate <the object
?
> to god.
6. SOME OTHER LATER INSCRIPTIONS
6.a. Stone of Trvik A
It is a stone of slate with a length of 234 cm, a width of 70 cm
and it is 8 cm thick. The runes have an height of ca. 11-15
cm and they are carved clearly. The inscription is not
compete because the stone was secondary smallered; the last
rune is cut through its staff. The stone is part of a grave and
was taken away from its original place of origin and was cut
to fit in the grave. Because the other stones are of a different
material it is clear the the inscription was formerly in a
different context than a grave-context.

The stone was found 1880. In the grave there was another stone
with a runic inscription (stone Trvik B) which is apparently
younger than that on Trvik A. The inscription cannot be
dated. The dating of the grave may not be taken over for the
inscription. The grave is dated between the 4
th
and 6
th

century. The form of the j-rune was seen as a proof that the
inscription has to be dated in the early 5
th
century. Of course
an earlier dating is possible.


http://www.arild-hauge.com/arild-hauge/no-rune-toervika-a.jpg
I
?

II ()

5

10

I u
?

II ladawarijaz

5

10

Interpretation:
la<n>dawarijaz is a nom.sg. masc. a-stem, dithematic personal
name < protogerm. *landa ari a-, as a name continued in
lgb. Landoari, ohg. Lantwari, Lantweri; a compostion of
la<n>da- and -warijaz.
la<n>da- < protogerm. *landa- land > goth. land, ohg. lant,
osax., oe., ofris., oisl. land; cp. nswed. dial. linda fallow
ground (< protogerm. *lend-), oisl. lundr groove (<
protogerm. *lunda-).
Most plausible interpretation: defender of the land.

The sign in line I has the form of an u-rune. But the function of
the sign cannot be secured. The assumption that it is a later
added u-rune or a bracket which connects the end of the first
element with the second one is not probable. It could be
thought that it was in fact the beginning of the inscription.
The carver wanted to write a l-rune but slipped away. After
that (because the sign could not be read as a l-rune anymore)
the carver decided to write the l-rune again more below.
Because the inscription is broken of it cannot be decided
whether there was a verb.
6.b. Ring von Strrup
The object is best referred to as a ring because it cannot be
decided whether it is a diadem, a neck ring, or even
something else. The ring is of gold, it weighs 405,6 gr and
has a diameter of 191-202 mm. The whole object was round
but the front part is made flat so it reaches an height of ca.
23,97 mm. The inscription is on the inside. The outer side is
stamped with ornaments. The runic signs have an height of
ca. 8 mm. The ring was discovered in 1840 in a grave. The
inscription is usually dated about 400 without any arguments.
In principle a time span from 160 400 seems to be possible.



()

5

lero

5
Interpretation:
lero is a nom.sg. masc./fem. n-stem, a nickname < protogerm.
*leran-/n- the leathern one, a derivation with the
individualizing suffix protogerm. *-n- of protogerm. *lera-
skin, leather > ohg. ledar, mlg. led(d)er, oe. leer, ofris. lether,
leder, oisl. ler.

6.c. Wooden box of Garblle (Stenmagle)
The totally unornamented wooden box has a length of 17 cm, a
width of 5 cm and an height of 5 cm. It was made of one
piece of wood of the yew tree and can be closed with a
sliding lid. Similar boxes were used to retain jewellery or
coins. The runic inscription stands when the lid is upside
in the false direction. The empty box was found 1947. A
dating is not possible. Because of the runes a dating around
400 was brought forth but every dating between 200 and 550
seems to be possible.

http://runer.natmus.dk//userfiles/image/ImgCache/614/Garb%C3%B8lle.JPG
()

5 10 15

hagiradaz tawide

5 10 15
Remarks: The division marks consist out of five dots but the dots
of the first mark build a connected row now. The assumption that
the first division mark must be understood as an i-rune has to be
given up in light of new investigations.
Interpretation:
Two-word inscription directly to be compared to shield mount 2 of
Illerup dal (niijo tawide), so with a sequence of a personal
name and a verb.
hagiradaz is a nom.sg. masc. a-stem, dithematic personal name <
protogerm. *aira-, a composition of hagi- und -radaz.
hagi- is an i-stem adj. < protogerm. *ai- apt, appropriate; cp. mhg. -
hac, oisl. hagr < protogerm. *aa- apt, complaisant; oisl. hgr <
protogerm. *i- apt; easy, complaisant; a derivation with the
verbal adjectives forming suffix protogerm. *-i- of *a/e/a- strong
verb VI to make apt > ohg. [part.prt.] kehain nurtured, mhg.
(part.prt.) behagen fresh, friendly; cp. osax. -hagon, oe. -hagian,
ofris. hagia, oisl. haga < protogerm. *a e/a-.
The element protogerm. *ai- appears only in personal names.
-radaz < protogerm. *ra- counsel > ohg. rt, osax. rd, oe.
rd, ofris. rd, oisl. r, a derivation with the verbal nouns
formative suffix protogerm. *-a- of protogerm. *re/a- to
counsel, to rule > goth. -redan, ohg. rtan, osax. rdan, oe.
rdan, ofris. rdan, oisl. ra.
This word is attested widely in personal names. The composition
has a counterpart in oisl. hagrr only that this word continues
protogerm. *aa- fort.
As is the case with the shield mount 2 of Illerup dal (niijo
tawide) it is unclear whether the name mentions the producer
of the object or the carver of the inscription.
Most plausible interpretation: Hagiradaz (= apt-counsel) made
[the object/the inscription].
6.d. Stone of Amla
It is a nearly square granitic with a length of 2,30 m, a width of
0,83-0,88 m and it is 0,21-0,33 m thick. The inscription is
not complete and shows a length of 1,02 m and an average
height of 15 cm. It cannot be decided how much broke away.
The inscription was found 1903. The dating of the immoble
object is difficult. The usual dating in the second half of the
5th century should be questioned because in the inscription
the diphthong ai is preserved before w so it could be older.


http://www.arild-hauge.com/arild-
hauge/no-rune-amle.jpg
()

5

10

izhaiwidazar

5

10
Remarks: Before the i-rune a half staff of another rune can be
seen but not securely identified. Although the inscription bears no
word division marks the partition in words is not difficult.

Interpretation:
iz is a nom.sg. masc. i-stem, most probably of a personal
name.
haiwidaz must be interpreted as a miscarving for hlaiwidaz
which is a nom.sg.masc. part.pret. of the verb protogerm.
*la e/a- weak verb I bury (not continued), a derivation
with the denominal nouns forming suffix protogerm. *- e/a-
of protogerm. *la a- grave > run. (nom.sg.) hlaiwa (stone
of B), goth. hlaiw, ohg. hlo, osax. hlo, oe. hlw, hlw.
The interpretation of h as a bind-rune hl seems not possible.
Interpretation:
iz is a nom.sg. masc. i-stem, most probably of a personal
name.
haiwidaz must be interpreted as a miscarving for hlaiwidaz
which is a nom.sg.masc. part.pret. of the verb protogerm.
*la e/a- weak verb I bury (not continued), a derivation
with the denominal nouns forming suffix protogerm. *- e/a-
of protogerm. *la a- grave > run. (nom.sg.) hlaiwa (stone
of B), goth. hlaiw, ohg. hlo, osax. hlo, oe. hlw, hlw.
The interpretation of h as a bind-rune hl seems not possible.
ar < protogerm. *ar there (partly with lengthening in
onesyllabics) > goth. ar, ohg. dar, osax. thar, oe. ar, r,
ofris. thr, oisl. ar, oswed. ar, r.
The often found rendering of as ar here seems not possible.
Most plausible interpretation: iz (is) buried there.
6.d. Goldhorn B of Gallehus
The goldhorn consists of two layers, an inner one consisting out
of a gold-silver alloy and an outer of gold. The horn is
ornamented heavily. The upper hoop bears the inscription.
The object has a curved form. The former function of the
horn can only be guessed: drinking vessel, music instrument
or offering gift. The object is not preserved but after it was
stolen 1802 it was melted immediatly. Two replica are lost
too. There exist now only drawings of the original object
from Krysing 1734, Paulli 1734 and Frost (in Gutacker)
1736. The dating around 400 it because of the ornaments.


ache/11/rune02b_305.
jpg

http://www.arild-hauge.com/arild-hauge/de-rune-gallehus.jpg
()

5

10

15

20

25

30
ekhlewagastiz:holtijaz:horna:tawido:

5

10

15

20

25

30

Remarks: The difference between the carving style has to be
explained by space problems.
Interpretation:
ek is the nom.sg. personal pronoun 1
st
sg. I.
hlewagastiz: nom.sg. masc. i-stem, dithematic personal name <
protogerm. *le aastiz, a composition out of the elements
hlewa- and -gastiz:
hlewa- either < protogerm. *le a- famous; cp. goth. hliuma
hearing, oisl. hljmr sound, noise < *le man-; oisl. hlj
hearing < *le /a-; ofranc. personal name Chlodo-, ohg.
Hludu-, oe. Hloth- < *luu/a-; perhaps also lat.-germ.
personal name (nom.sg.) Chlevvia.
In this interpretion the name could be compared with gr.
personal name .
or < urgerm. *le a- protection, protecting place > run.
(nom./dat.sg.) hli (clasp of Strand, ca. 700), perhaps (acc.sg.)
he (stone of Stentoften, ca. 650), mhg. lie, osax. hleo, oe.
hlo(w), ofris. (acc.sg.) hli, oisl. Hl.
Although this interpretation seems to be less likely it can be
defended ragarding the oe. Junction gesta hleo protection of
the guests.
-gastiz < protogerm. *asti- guest > goth. gasts, ohg., osax.
gast, oe. gst, giest, ofris. jest, oisl. gestr; occuring often as
an element in personal names, cp. lat.-germ. Arvagastis,
Arvagastes, Arbogast, Harigasti.
holtijaz is a nom.sg. m. a-stem < protogerm. *ulti a- the one
belonging to *hulta-, a derivation with the affiliation
forming suffix protogerm. *- a- of protogerm. *ulta- wood
> ohg. holz, os., olfrk.., oe., ofris., oisl. holt.
The result of th a-Umlaug (u > o before a in the next syllable)
had already phonemic status because the precondition is not
at hand; it must have been introduced from the basic word
*ulta-.
horna is an acc.sg. neutr. a-stem < protogerm. *urna- horn >
run. (nom.sg.) horna (whetzstein of Strm; 5. Jh.), goth.
haurn, ohg., osax., oe., ofris., oisl. horn.
tawido is a 1.sg.prt.ind. of the verb *ta e/a- weak verb I do,
make (cp. tawide mount shield 2 of Illerup).
The syntactical connection between the personal pronoun and
the personal name seems to be marked by the absence of a
word division mark.
Metrically this is the oldest document for stave rhyme in
Germanic: ek l astiz lti az | rna t io.
Most plausible interpretation: I Hlewagastiz
(famous/protection-guest), the one belonging to *hulta, made
the horn .

Introduction To The Older Runic Inscription

Uploaded by

Copyright:

Available Formats

Introduction To The Older Runic Inscription

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Introduction To The Older Runic Inscription

Uploaded by

Copyright:

Available Formats

Roland Schuhmann JSIEL 2009 The early Runic Inscriptions

inebriant, beer > mhg. al- (in

and an analogy is not

and an analogy is not

You might also like