The Celts and the Origin of the Runic Script, 143-55.pdf

Studia Neophilologica 71: 143–155, 1999

The Celts and the Origin of the Runic Script

BERNARD MEES

The Germanic runes are usually stated to have their origin in an adaptation of a

Mediterranean script: Etruscan, Greek or Latin. 1 A Latin origin for the runic script (known

as the fuþark, after the first six characters of the runic script as represented on the Kylver

stone), as was most influentially formally propounded by Ludvig Wimmer, 2 suggested the

likelihood that the runes were adapted or evolved from the Latin ABC through mercantile

contacts in the region of the Rhenish or Danubian frontier. A similar method of adoption

from an eastern Greek alphabet, as was subsequently suggested by Sophus Bugge, has lost

favour owing to the emphasis of Bugge and Otto von Friesen 3 on the Black Sea Goths as

the tribe most likely to have adopted the characters of the Greek world. The Etruscan or

rather the descendent North Italic alphabets, as was suggested by Carl Marstrander and

Magnus Hammarstr ¨m, 4 are often considered to be more similar orthographically to the

runic script, and boustrophedon and sinistroverse runic inscriptions also point towards a

North Italic influence as many of the Italic inscriptions are executed from right to left. A

convincing historical argument as to the date and method of adoption of the fuþark,

however, has yet to appear. The North Italic thesis is hampered by the fact that the first

runic finds in the south of Germania appear later than do those of the north, not dating

before the fifth century A.D. 5 Marstrander and before him Holger Pedersen suggested that

Celtic tribes might have acted as intermediaries in the centuries about the birth of Christ,

whether from North Italic or Roman sources. Yet although the North Italic thesis once

received widespread support among runologists, the only widely accepted attempt to

explain the agency of transmission granted a North Italic source was influenced by the

Negau B inscription and the idea of Alpengermanen in the second century B.C., generally

regarded now as a myth. 6

Another theory that has endeavoured to explain the transmission of writing from south

to north is the Cimbrian thesis that seeks to link the acquisition of the runic script with the

return north of defeated Cimbri and Teutones after the Roman victory at the Vercellae in

101 B.C. 7 Although linked in its original proposition to the North Italic thesis, it could also

be applied to an origin theory founded on any other script used in a region through which

the Cimbri wandered. The southward journeying of these people, probably from northern

Jutland, is often linked to importation from the south as a reflux movement enabled by the

opening of the southern frontier. 8 The departure of the Cimbri is an early turning point in

Germanic history, as it first introduced the furor Teutonicus to the Mediterranean region,

but contact with the south had been occurring since the Bronze Age, and despite a

contraction in trade at the time of the fall of the Late Halstatt F¨ rstensitze , north-south

links had been reopened before the time of the departure of the Cimbri. 9 Early Greek

explorers such as Pytheas had visited or heard of Germanic tribes two centuries prior to the

southern irruption of the Cimbri, although it was not until the appearance of the

ethnography of Poseidonios that the term Germani was applied to these people. 10

Cimbrian theses are merely historical crutches used to provide a defined event to illustrate

the beginning of many different north-south relations that were already burgeoning before

the Cimbri irrupted so violently into Mediterranean history.

Until the discovery of the Meldorf fibula inscription in 1979, the earliest of the over 200

runic finds in the older fuþark were usually dated to the late second century A.D., a century

before the Goths arrived at the Black Sea shores. Even so, datings for early finds such as

the Øvre Stabu spearhead (unearthed in 1890) were somewhat controversial until a proper

absolute (vis-`-vis relative) archaeological chronology for the Iron Age was developed

144

B. Mees

Studia Neophil 71 (1999)

after the war, a lemma that explains why the thesis of a Gothic adoption of Greek letters

was maintained by Gunnar Ekholm, a prominent Scandinavian archaeologist, as late as

1959. 11 A modified eastern thesis replacing the Goths with the earlier eastern Germanic

Bastarnae or Quadi remains a possibility, however, and Greek letters were in use in Gaul

from the third to first centuries B.C. Nevertheless, von Friesen’s Greek theory relied on

cursive Greek, not the epigraphical letters typical of pre-Roman Gaul. Yet a Greek origin

is still maintained by runologists who seek runic origins in pre-classical times, and it is

clear that many imports into the Germanic North come from eastern sources, as is

evidenced by finds such as the famous Thraco-Celtic cauldron from Gundestrup and the

standard measure of Borremose which is a Greek foot (33 cm), not Roman (29.5 cm). 12

The inscription on the Meldorf fibula has been dated on stylistic grounds to the first half

of the first century A.D. Although the interpretations proffered for this inscription are

mostly problematic, most runologists see Meldorf as at least proto-runic, even while

maintaining different theories as to the origin of the runes. The Meldorf inscription cannot

be semantically Latin, and Germanic inscriptions in Roman letters are unparallelled for

this period, the only roughly contemporary inscriptions being the orthographically North

Italic Negau B helmet inscription, and the doubtful evidence supplied by an ortho-

graphically Latin coin legend from Pannonia. (The earliest Germano-Latin inscriptions

attributed to the Rhenish Ubii date from the second–third centuries A.D.). Although it has

not provided any orthographic clues as to the prototype for the fuþark, the discovery of the

inscription on the Meldorf fibula has probably pushed back the date of the earliest use of

the runes by the Germanic peoples to before the time of Tacitus. Indeed another first

century find has recently come to light in excavations from Osterr ¨nfeld, less than 50 km

northeast from Meldorf, that bears an even briefer inscription that may also be runic. 13

Many runologists suppose that the fuþark was borrowed or created as an indigenous

response to the need for a system of writing for use in mercantile contracts such as those

employed by Mediterranean traders. The underlying assumption here is that the Germanic

peoples traded according to the rules of the Mediterranean cultures, a notion not borne out

by modern archaeological constructions. Trade during the Germanic pre-Roman Iron Age

was more inclined to diffusive tendencies or was dominated by ‘big men’ or chiefs in the

modern reconstructions based on anthropological models. The evolutionary models of this

archaeology indicate a type of economic structure quite different from the market

economies of central Europe and the Mediterranean, although since the time of Caesar

these economic systems had increasingly come into contact. The concept of a Germanic

mercantile class is not borne out by historical or archaeological evidence. 14 The notion of a

mercantile need for the fuþark is obviously influenced by the written testaments of many

other languages, such as Linear B Greek, that are records of traded commodities. The

mercantile use of the fuþark, however, is not evidenced by any ancient (pre- V¨ lker-

wanderung ) inscriptional find. 15

Others suggest that Germanic warriors may have brought the Latin letters back with

them after serving in the Roman army. This thesis is predicated on archaeological

groundwork that indicates there is a concentration of Roman artefacts on Jutland and

Sjælland, finds that are not parallelled closer to the borders of the Empire. As the earliest

runic inscriptions also stem from this area, the same dynamic that produced the pattern in

the distribution of imported artefacts is assumed to have affected the distribution of the

early use of runes. Yet this pattern is only apparent from the fourth century A.D. when the

archaeological record indicates centralising forces that are assumed to reflect the

development of a centralised power on Sjælland, i.e. the early Danish state. Thus the

runes are thought to have been adapted from Latin by warriors returning from Roman

service for diplomatic, social or legal purposes in this early Danish state. The earliest runic

finds, however, clearly pre-date the period for this reconstruction by at least two centuries.

Although Roman goods first appear in this region from the late second century, at this

earlier time it is clear from historical sources that Germanic mercenaries mainly came

The Celts and the Origin of the Runic Script

145

Studia Neophil 71 (1999)

from the regions closest to the Empire, not Scandinavia. Indeed figures such as Arminius

would clearly have been literate, and Germanic soldiers had probably served in the

retinues of Celtic leaders in the last century B.C. or even somewhat earlier. This thesis

clearly does not explain the distribution of the earliest runic finds. Moreover, although

many runic inscriptions are found on weapons and shields of Germanic manufacture, there

are none found on Roman imports. Thus imports of Roman manufacture are not directly

linked with runic inscriptions, though both may be linked with a growing emphasis on the

possession of prestige goods. Nevertheless, there is no evidence that indicates that a

Germanic warrior brought the prototype upon which the runes were based back with him

after a sojourn in the South. 16

A cultic origin of the runic script is less well received. Many of the earliest runic finds

are often characterised as pertaining to magico-religious discourses, and not surprisingly,

the names of some Germanic deities have been recognised in early Germanic forms. 17 The

most aggressive example of this approach is one that seeks to characterise the evolution of

the runic script as not from a Mediterranean prototype but rather as an autochthonous

response utilising symbols that had appeared in Germanic (and indeed European) art since

the Bronze Age. Of course the idea of orthographic transmission must have been borrowed

from the cultures of the Mediterranean or Near East, but the form that this orthography

took is, in this approach, held to be a Germanic innovation. 18 The creation of the script at

one time and at one place seems to be indicated by the standard form of the older fuþark,

with independent traditions appearing only after the V ¨ lkerwanderungszeit . It is also

extraordinary that the language of the older inscriptions, somewhat ahistorically termed

Primitive Norse, 19 is considered by some linguists as a koin´ , implying that these

inscriptions do not represent the dialects of Germanic of the time, but rather a standard

runic language. 20 The autochthonous theory is undermined by the similarities between

runic and the Mediterranean scripts in the form of many of the characters, the number of

characters in the rune-row, the regular alphabetic rather than syllabic or semisyllabic

values, the representation of vowels and the continued use of pre-runic ideographs (such as

the swastika) that are sometimes found embedded in runic inscriptions. Hence the search

for a Mediterranean prototype of the Germanic script.

There were a number of scripts in use in Italy other than Greek, Etruscan and Latin in

the last centuries B.C. such as the four North Italic Alpine scripts of Lugano, Sondrio,

Bolzano and Magr` and the related Venetic script of Este. The first of these, as was first

clearly shown in 1970, was used by Celtic peoples of the Golasecca culture and later La

T`ne Cisalpine Gallic invaders (who also adopted other Italic scripts such as Venetic and

Subpicenian). Further east, in Carinthia, another script was used by the Celtic Norici in a

region proximate to the northernmost Venetic finds. A modified Iberian script of the

Celtiberian peoples was employed in northern Spain and southern Gaul, and Celtic peoples

also used an Ionian Greek script borrowed from the Phocaean colonists at Massalia, only

adopting Latin letters after the creation of the Roman province of Gallia Narbonensis. 21

The discovery of the Meldorf fibula, even if we accept that it bears only early or proto-

runes, pushes back the time for the adoption of writing by the ancient Germans to at least

about the time of Christ. 22 Yet it is unlikely that this inscription represents the first native

Germanic attempt at orthographic representation. Such finds suggest, then, that the earliest

use of the runes pre-dates the Roman expansion to the Rhine and Danube. If this is the

case, then the most likely model for the runic script would be one of the scripts employed

by Celtic peoples in the last centuries B.C. noting the large amount of Celtic material

imports of this time which have been seen as indicative of the method of transmission of

much technology to the North. 23 Indeed certain letters of the Celtiberian script show

remarkable similarities in form to runic, as do those of the Lugano, Noric, Venetic, Latin

and Greek scripts used by these peoples.

Nevertheless, orthographically the runic script is held in the handbooks to be most

similar to the North Italic alphabets. The comparisons, however, rely on not one particular

146

B. Mees

Studia Neophil 71 (1999)

North Italic script, but rather an amalgam of the five mentioned above. This orthographic

preference is indicated in at least eleven of the North Italic characters, significantly more

than that of Latin or Greek. 24 Yet the number of North Italic correspondences is reduced

significantly if each of the scripts is considered separately, which makes the comparison

charts, most of which are copied from Marstrander’s original contribution, rather

misleading. Solutions to this problem deriving solely from one-to-one comparisons of the

runes with Mediterranean scripts have in the main been unconvincing. A more successful

approach may be to study other criteria for the origin of the runic script.

We know one of the methods of divination employed by the Germanic peoples of the

first century thanks to a description by Tacitus. This divinatory practice that talks of

written symbols is described in the following passage in his Germania of A.D. 98: 25

auspicia sortesque ut qui maxime observant. sortium consuetudo simplex. virgam frugiferae arbori decisam in

surculos amputant eosque notis quibusdam discretos super candidam vestam temere ac fortuito spargunt. mox, si

publice consultatur, sacerdos civitatis, sin privatim, ipse pater familae, precatus deos caelumque suspiciens, ter

singulos tollit, sublatos secundum impressam ante notam interpretatur.

Previously, under the influence of the autochthonous theory, these notae had been equated

with Germanic ideographs such as the swastika, triskelion, fulmen and tamgas. Yet from

the Classical Latin use of nota ‘indication, symbol, letter, cipher’, it seems that Tacitus, as

is his usual practice, has probably translated a Germanic expression. As we now know, due

to the discovery of the Meldorf fibula, that the Germanic peoples employed writing before

the time of Tacitus, this term can scarcely be any other than ‘rune’. What is probably our

earliest historical reference to Germanic writing is, therefore, connected with a cultic

discourse, an observation that agrees with interpretations of many of the early inscriptions

which stress their invocatory nature. 26

Many of the early inscriptions state that they were written by an erilaz , a word that

evades a convincing etymology. 27 One type of runic inscription that appears from an early

date is the setting out of the fuþark, often independent of any other message, a type of

inscription well known in other protoliterate societies such as that of the early Etruscans,

Latins, Greeks and Gauls. 28 Considering the connotation of mastery in the Germanic root

* er -, the erilaz is best interpreted as a master of the runes and it is possible that the rune-

row inscriptions were intended to indicate that the writer was proficient in the use of the

fuþark; that he or she had been initiated into its use. 29

Sometimes in these inscriptions the order of pairs of runestaves is reversed. 30 These

alternations are best explained by a mechanical system underlying the order of the fuþark:

i.e. it was a group of paired characters. Obviously, then, the rune masters learned the

runestaves in pairs, and sometimes the order of some pairs was confused.

We also know the magical implication of many of these early runic inscriptions owing

to the appearance of words such as alu (cf. Hitt. alwanza - ‘bewitched, ensorcelled’, Gk.

al´ ¯ ‘be beside oneself’, Lett. aluˆ t ‘wander aimlessly’; OE ealu ,OS alo -, ON ˛l ,PG

* aluþ - ‘intoxicating beverage’ > ‘ale’) that can only be described as invocatory. 31 A

notable example often promoted as representing a magical use of the runes is the

inscription on the fish-shaped bone piece from Lindholm. Not only does it include the

statement: 32

ek erila R sawilaga R ha(i)teka

‘I, the erila R am known as Sawilaga R ’,

but the reverse contains an inscription that seems completely meaningless:

aaaaaaaa RRR nnn

b m u ttt : alu :

Some authors have suggested that both inscriptions were construed as 24 characters

long, as many as there are in the fuþark. A similar connection between the inscription and

number of characters has been seen in other inscriptions. Many such arguments are surely

far-fetched, yet the sequence of repeated runes on the Lindholm piece seems to indicate

The Celts and the Origin of the Runic Script

147

Studia Neophil 71 (1999)

some sort of cryptological, or considering later magical practices, a magico-religious

expression. 33

It has also been observed by many linguists that the staves of the fuþark show

inadequate concern for the Germanic phonological system. The best known indication of

this is the appearance of the yew rune, which is usually represented by ¨ in the grammars,

the phone that it represented being a matter of some dispute, but it is generally held to be

identical to that of the ice rune, i.e. an /i(:)/. The retention of what seems to be a

phonetically redundant character in an adopted script is very odd, and many different

theories have been proffered to explain this aberration. 34

The name of each rune is also interesting for the question of the origin of the runes

because neither the Latin nor Greek scripts had meaningful acrophonic names for each

letter. It is also notable that some inscriptions suggest that the individual rune names were

known well enough to be used as ideographs rather than actually spelling out the lexeme so

represented. This suggests that the names may have been more than mere mnemonic

devices. 35

It is noteworthy that a seemingly unconnected, lexically meaningless inscription has

come to light in Pompeii. This inscription is a series of Latin letters in pairs, the first letter

of each pairing starting with A and heading forward in the ABC, the second with X and

heading backward: 36

AXBVCTDSER

The Pompeii inscription has been linked by Elmar Seebold to a practice known in Near

Eastern societies, divination by letters, especially in the form known to the Semites as ath-

bash . 37 This system relied on pairs of letters grouped from alternate halves of the script, as

occurs in the Pompeii inscription. If the order of the Germanic rune-row is compared to

that of the Mediterranean scripts a similar pattern can be discerned. The sound values of

the fuþark when compared to the Mediterranean scripts indicate an ath-bash or a similar

mechanical pairing system, which, if Seebold is correct, was known to the Mediterranean

cultures in the centuries B.C. 38

This is remarkable given the nature of the rune names recorded in medieval sources,

from which scholars have, through the comparative method, produced a series of Common

or Proto-Germanic names, one for each character of the older fuþark. The names seem to

appear mostly in semantic pairs and the ideas represented by these rune names can nearly

all be construed as pairs of complements or oppositions (e.g. the names of f and u are * fehu

and * ¯ ruz , ‘cow, livestock, wealth’ and ‘aurochs’, tamed and untamed; þ and a are

* þurisaz and * ansuz , ‘giant’ and ‘god’, the cosmic opposites of the mythological world). 39

Additionally, the rune pairs often seem to be semantically linked to other pairs. This

observation has led some runologists to make the controversial suggestion that the rune

names can be collected into three semantic groups of eight with the conceptual meanings

‘living thing’, ‘nature’, and ‘society’, groups that mirror the separation of the staves into

three eights or families (ON ættir ) in medieval sources. 40 This proposed system seems no

accident, and given the possible influence of an ath-bash or a similar mechanical

divinatory system as indicated by the paired order of the rune row, and the presence of

phonological inconsistencies such as the yew rune, the evidence points to an early

utilisation of the runic script in an oracular system. Thus it may be possible to rediscover

something of what Tacitus means when he speaks of divination by reading the cast

notae . 41

Yet is there a direct connection between Mediterranean and northern uses of divinatory

writing? Seebold has not been able to demonstrate adequately a numerical system upon

which the rune-row is based that corresponds to another oracular system which would

provide a prototype for the underlying system of paired names and order of the fuþark,

although he has indicated mechanical parallels. 42 . Seebold believes that there is a direct

influence in the mantic use of writing among the semiliterate in the Mediterranean on the

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: