Uploaded by oohay.skcus

A Celtic script in the eastern La Tène culture ecelt 0373-1928 2003 num 35 1 2152

Etudes Celtiques
A Celtic script in the eastern La Tène culture ?
Jürgen Zeidler
Citer ce document / Cite this document :
Zeidler Jürgen. A Celtic script in the eastern La Tène culture ?. In: Etudes Celtiques, vol. 35, 2003. pp. 69-132;
doi : https://doi.org/10.3406/ecelt.2003.2152
Fichier pdf généré le 07/05/2020
Une écriture celtique dans la culture laténienne orientale ?
Dans une vaste zone de la culture laténienne orientale, s’étendant sur l’axe ouest-est de la Bavière à
la Hongrie occidentale et sur l’axe nord-sud de la Bohême à la Carinthie et la Styrie, des récipients
graphités ornés au peigne ont été découverts, datant surtout du premier siècle av. J.-C., sur le fond
desquels apparaissent divers signes. On en connaît aujourd’hui environ 1200, dont à peu près 660
sont rassemblés ici dans un appendice (à l’exclusion des matériaux noriques). Jusqu’aujourd’hui, ces
signes ont été généralement pris pour des marques individuelles de potiers. Cependant, l’auteur
propose ici la théorie que ces signes révèlent un alphabet latènien presque complet. De brèves
inscriptions sur des céramiques, sur des monnaies et, dans un cas, sur un bois de mine montrent les
mêmes formes de lettres et donnent quelques indications sur leur valeur phonétique. Des lectures
prudentes semblent indiquer des noms et des mots en gaulois. Notre argumentation est que l’écriture à
la base des signes fut adoptée et adaptée au plus tard au troisième siècle av. J.-C., c.-à-d. à l’époque
des premières monnaies. L’alphabet, selon toute vraisemblance, dérive du camunique ou du vénète,
de même que les écritures régionales ‘rhétiques’ et le norique sur le Magdalensberg. L’emploi de ce
système d’écriture s’est maintenu jusqu’aux premiers siècles apr. J.-C. et fut évidemment adopté par
un peuple avec une culture matérielle ‘germanique’. Des signes similaires sur des pots existent
apparemment au début de la période romaine en Moravie et sporadiquement en Silésie, Brandenburg
et Hesse. L’écriture laténienne montre aussi une similarité remarquable avec le futhark runique : elle
pourrait avoir été un stade intermédiaire dans la transmission d’un type (en fin de compte) étrusque de
l’alphabet en Allemagne du Nord et en Scandinavie où les premières inscriptions ‘proto-runiques’ et
runiques ont été découvertes.
Jürgen Zeidler : A Celtic script in the eastern La Tène culture ?, p. 69-132.
In a large area of the eastern La Tène culture, extending West-East from Bavaria to western Hungary
and North-South from Bohemia to Carinthia and Styria, comb decorated graphite vessels have been
found, mostly dating from the first century BC, on which a variety of marks appear on the bottom. About
1200 of them are known up to now, and c. 660 are listed here in an appendix (with the exception of
Norican material). Previously, these vessel marks have usually been understood as individual potters’
brands. In this paper, however, the theory is proposed that the marks reveal an almost complete La
Tene alphabet. Short inscriptions on vessels, coins and in one case on mine timber show the same
shapes of letters and give some hints at their phonetic values. Tentative readings seem to point to
names and words in the Gaulish language. It is argued that the script underlying the marks was
adopted and adapted in the third century BC at the latest, i.e. the time of the earliest coins. The
alphabet is in all probability derived from Camunic or from Venetic, like the ‘Raetic’ local scripts and
Norican on Magdalensberg. The use of this writing system was maintained until the first centuries ad
and obviously adopted by people with a ‘Germanic’ material culture. This is clear from similar vessel
marks of the early Roman period from Moravia and sporadically from Silesia, Brandenburg and Hessia.
The La Tène script shows a remarkable similarity to the runic Futhark as well, and it is suggested that it
may have been a Celtic intermediary stage in the transmission of an (ultimately) Etruscan type of
alphabet to northern Germany and Scandinavia where the earliest ‘proto-runic’ and runic inscriptions
have been found.
In recent years the study of ancient Celtic scripts and languages has received a
great impulse since much new evidence has come to light. Most of the work concen¬
trates on Celtiberian, Gaulish in the Greek and Latin scripts, Lepontic, as well as
adjoining languages like Yenetic and ‘Raetic’ (on the inaccuracy of the term, see
Schumacher 1998: 334 f). In my opinion, a further, if minor, source can be added to
the spectrum which has not yet received the attention it probably deserves. The
reason is that the study of the relevant material seems to be confined to the domain of
prehistoric archaeologists although it could also be interesting to historians of writ¬
ing, linguists, and celticists in general. The ‘Celtic’ character of the script is not evi¬
dent from the beginning but can be confirmed by several observations.
The main body of evidence under consideration here is made up of markings on
late eastern La Tène ceramics. Some vessels probably date back to the period of late
middle La Tène (C2), c. 200-120 BC, but most of them from late La Tene (Dl-2), c.
120-50 and 50—10/0 BC according to traditional chronology. Higher absolute dates
for Dl/2, c. 150-80 / 80-10 BC, suggested by S. Rieckhoff (1995: 194), are not
accepted by E Fischer (1999: 383 f) and others. A few markings come from the early
Roman periods (Bl-2, Eggers 1955), c. AD 0-50 and 50-150. Single marks may have
been known in parts of Austria in late antiquity according to G. Moßler (see Eichner
1994: 138; and e.g. Hahnel 1994: 56, fig. 36). The signs appear on the exterior side
of the bottom of large open bowls which are, as a rule, made of graphite clay or clay
with a graphite component. Pots of this sort were highly esteemed because of their
heat-resistence, hydrophobic character and impact resistence (Jerem & Kardos
1985: 65; Waldhauser 1992: 378) which made them particularly suitable for cooking
over an open fire; because of these qualities, graphite vessels were still in use in the
late Middle Ages (e.g. Cech 1989: 168; in general Camap-Bomheim 1998). Usually
the vessels are decorated with multiple parallel incisions produced by a comb
(Kammstrich). It is remarkable that the marks have been carved with the finger or a
small rod into the moist clay before the pottery was burnt (ante cocturam). Only a few
of them have been incised after burning (post cocturam ). Markings of this kind are
known as ‘bottom signs’ (Bodenzeichen) for at least eighty years since M. Hell (1922)
published the first one in Wiener Prähistorische Zeitschrift. Today more than 1200
marks on ceramics are known and many more may still lie unrecognized in archaeo¬
logical collections. It must be noted, however, that a large portion of the material
comes from a few large late La Tène central places, in particular Manching (c. 200
marks), Hallstatt (c. 120), Stare Hradisko (c. 60) and the pre-and early Roman tra¬
ding centre Magdalensberg (c. 70, together with some other sites, see Hebert 1991:
285 fn. 21; Zabehlicky-Scheffenegger 1997: 132; ead. & Sauer 1998).
In southern Noricum and the Middle Danube, tripod bowls with a lid made of
coarse grey clay also show vessel marks (Schindler-Kaudelka & Zabehlicky-Schef¬
fenegger 1995; Zabehlicky-Scheffenegger 1997; on Gleisdorf: Artner 1988/89). Most
of the c. 580 marks known today come from Magdalensberg and date from the early
Roman period, early Augustan to Claudian. Like graphite vessels, tripod bowls were
used as cooking dishes, the lid preventing heat loss. Their capacity ranged from 0.8
to 2.7 litres. Vessel marks were incised ante cocturam on one of the feet (39%), on the
side (8%) or on the lid (53%). Signs on the base and the side give a hint at the orien¬
tation of the marks, which is not obvious from lid or bottom incisions.
A minor part of the material is made up of similar markings on metal objects, par¬
ticularly buckets and jugs originating from Etruria or imitating Etruscan products.
Their range of dating is from c. 1st c. BC to 1st c. AD. In this case, the marks are
always incised into the exterior upper part of the receptacles. Probably the first one of
these was published by M. Jahn in 1919. Incisions on metal are already known from
the Hallstatt D period (5th c. BC) as e.g. on the famous cult waggon from Strettweg,
Styria and other objects (Egg 1996: 53-61; Mayer 1976). Recently an inscription on
a finger ring from Nußdorf (Chiemsee, Bavaria) has been published (Rix 1998) which
was dated to the 5th c. BC. Unfortunately it cannot be attributed to a specific period
since it is not at all clear whether the ring and datable finds were deposited at the
same time. Unlike ceramic marks, there have always been but a few marks on metal.
Since not all signs can be regarded as evidence of writing (Mayer 1976: 378 f), they
cannot be fully included in this study.
A further part of the material is made up of similar signs on Celtic coins dating
from the mid 3rd — 1st c. BC. They are not universally acknowledged as letters —
most numismatists (e.g. Kellner 1990) describe them as ‘strokes’ or ‘symbols’ — so
one has to be cautious to include them into the corpus. But several signs are so simi¬
lar to vessel marks that it seems justified to take them into account. Marks on coins
and other metal ware are a special case anyway and should be considered in a study
of its own; on legends in general, see Colbert de Beaulieu (1956; 1965; 1965a),
Marinetti & Prosdocimi (1994), Arslan (2000), RIG IV. 11-15. The following coin
types show comparable markings: (1) Golden rainbow cups (Regenbogenschiisselchen), especially bird head staters (Vogelkopfstatere ) type II D, perhaps II E
(referring to the tables in Kellner 1990); large quantities of signs have been found in
the hoards from Bochum and Mardorf (Forrer 1910; Kappel 1976), but they are also
present in Bavaria, e.g. in Gaggers, Lkr. Dachau (Kellner 1990: pis. 49 f, nos. 1961—
1962) and Stöffling-Bedaium, Lkr. Traunstein (Kellner 1990: pi. 45, no. 1656; Egger
in Keltjt., 367, no. 520a). Their range of date is within La Tène D (120—10 BC). Ear¬
lier hoards from LT C2 (200—120 BC) hardly show any such markings, as e.g. Gro߬
bissendorf (Ziegaus 1995: 117-126). (2) ‘Boian gold staters of the older coinage’
(Castelin 1965) display similar marks in a recently discovered treasure of c. 450
coins in Manching (Leicht & Ziegaus 2000). Nike gold staters from the beginning of
the Bohemian coinage (c. mid 3rd c. BC, see now Ziegaus 1997) sometimes have
legends in native script, such as TELLO (De la Tour, pi. 40, no. 9474) though some
of them, e.g. ‘LIKI/L’ and ‘CIECINN/M’ (Forrer I, 195 f; DAG 1181 1), are problematic.
(3) Silver cross quinarii of the types Diihren and Schönaich may also be considered
to belong to this group because they show varying signs in the cross corners, proba¬
bly deriving from the legend VOLC of the Volcae-Tectosages, but some of them
resemble vessel marks (e.g. Kellner 1990: pi. 2, nos. 36—41; pi. 6, nos. 106 f; pi. 13,
nos. 277 f; 281-85).
East Norican silver tetradrachms with a ‘globe rider’ (Kugelreiter ), type C2K (Gobi
1989), probably minted on or near Magdalensberg and dated to c. 60 BC, bear Venetic
legends such as VES·, IEI KR and TI (De la Tour, pi. 51, nos. 9913-9916; DAG 1182 f) which
show some resemblence to Norican inscriptions. Strange to say, the sinistroverse legend
has repeatedly been misread after Forrer though the script has always been recognized
as (basically) Venetic. As the F-like letter is <v>, not <f>, the reading is <ves> (or <veso>
if the dot is <o>) which points to Venetic or Illyrian ves-or Celtic vesu-, veso-‘good, rich’
being a well known first and second element in personal names (KGPN 293 f; Lejeune
1974: 254 no. 138) as e.g. the famous Bellovesus and Segovesus (Livy, v. 34).
The theta and o-signs IS, Θ, Φ; O (resembling types F2, H3, H5; HI below) on Gaulish
(RIG IV. 11-15) and eastern Celtic coins (Gobi 1973: 28 f; pi. 47) usually must be left
aside because they may derive from the wheel of a biga on Macedonian coins and thus
have ornamental character, or from the legend of the Thasos type (ΘΑΣΙΩΝ), both of
which have been models of Celtic coinage; on Θ in context with letters, see below.
In addition, unique incisions on mining timber from Hallstatt may be taken into
consideration because they show similar signs and date from the 3rd to the middle of
the 2nd c. BC (Barth 1984). They probably provide the oldest evidence of the script
under investigation (see below, ‘fl p. 91-92).
In this study about 660 records on ceramics are listed (see end of paper), most of
them being already published. Some records, however, may have been overlooked.
Only a few new instances could be added so far (e.g. Berching-Pollanten; Dobrichov;
Kelheim). Material from Magdalensberg and other Carinthian and Styrian sites is
only touched upon because it will be published at large by E. Schindler-Kaudelka
and S. Zabehlicky-Scheffenegger (cf. Eichner 1994: 138; Moßler 1961; 1986). Marks
on metal objects and coins could not be collected systematically up to now and so
have not been fully included here.
There is a methodological problem in recognizing possible testimonies from the Roman
period because many markings resemble Latin letters or ciphers very closely, as X, V, I etc.
As these symbols often occur on Roman sites (see e.g. Egger 1959; Bakker 1975; Struck
C6 B5
algmdaelbenfsbure kg,t
Wat ens, okr.
Krumlov In sbruck
B3 B5 C5 D3 C7 C6
Engelsberg ase-Riesgase;
gr(Raomaf iton)
(Venetic lead
Pravcice, Pteni,
66. 67. RaD4
C6 B7
inscription) (‘Raetic’
Heilbrun er
Stut gart on
coarse and
D5 C4 C6 B5 B7
69. 70.
Radkersburg Roseldorf, Salzburg, Rainberg Stammheim,
Hol abmn
E F StD4
Gleisdorf Ljubljana Verdun Grafenstein Gurina Kalsdorf, iMagrè
(n‘E6DRscraipetioncs’) Manching Ptuj
D6 E5 E6 D6 E5 F5 E5 E4 E6 F3 C4 E6 D4
Jicin grey
(Cil i)-Celeia
Zliv, Coarse Bratislava Célje
D6 D6 C7 C7 B5 B6
C7 B5 C7
C7 C7 B5 C7 C6
Pistorf Polesovice, Postoupky,
92. 93. DieE5
tman sdorf 95. 96. 97. A B C
79. 80. 81. 82. 83. 84. 85. 86.
Breclav Hörsching,
Breclav okr.
C6 C6 C6 C4
C3 D5 C6 C7 C6 C6 C6 D5 C6 D7 C6 D4
87. 88. 89.
Strigova Tauberbischofsheim Trebusice, Irisov Volders Vracov, Waltersdorf Wien, Rudolfstiftung; Wien, Zahnasovice, Zalkovice,
Straubing Strebersdorf,
Oberpul endorf
C6 B5 C6 B5 C4 D6
SC5tradonice 76. 77.
73. StE5
74.rachotin, 75.
Hradisko, Prostëjov
Staré okr. Stochov,
C6 C6 B7
Manching, Micheldorf, okr.
Milocvlicaev, Mistrin, Moravské Musov, Nejdek, Neubau, Nëmcicky, Nitriansky
49. 50. 51. 52. 53. 54. 55.
E6 E4 D4 D5
raßundsotrefin, Horn
Oberthümau, Bez. Oslavany, Pavlov,
Linz, Lovcicky, Lysá
46. 47.
B6 C6 C6 D4 C4 D7 D3 okr.
42. 43. 44.
C6 E5
Ladendorf, Lanzhot, okr.
38. 39.
D4 C6 C7
Pfaf enberg 62. 63. 64.
59. 60. D5
B3 D4
Jicin of
Kolokr.in Brecl
Breclav okr.
Gänserndorf Bez.
okr.av Bez.
Blansko Kromëriz
Del ach Dammwiese, Christnastolen Chvalcov, okr.
Köfering, (?)
Kromerfz Komámo
Bratislav -mesto
Sa lfelden, okr.
Ainring, Reichenhal
marks quantities
Altendorf, Au, Bad Berching-Polanten Biberg, Boritov, Brest, Budapest-Gelérthegy-Tabán Bykovice, Deutschlandsberg, Devin, Dobrichov, Dmholec, Dürnkrut, Dürmberg, Egglfing, Farchant Frauenstein, Großrußbach Grub, Gurina Hal ein Hal stat Hal stat Hostyn, Hrazany, Hulin, Iza, Jaroslavice, Jicina Karlstein, Katsch, Kelheim Klentnice, Knëzeves, Koprivnice,
18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. •
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
1996: 83 f), one cannot be sure about their Latin or native origin. The latter can only be
confirmed where definitely non-Latin marks (or inscriptions) have been found.
The core dispersal area of the vessel marks extends from west to east from Bava¬
ria to Hungary, and from north to south from northern Bavaria and Bohemia to Carinthia and Styria, i.e. it comprises the main territory of the eastern La Tène culture (fig.
1). Sporadical finds are known from Württemberg, Brandenburg and Silesia. The
marks on coins enlarge this zone as far as Luxemburg and the Netherlands (cf. D.
Wigg 1999: 105). No similar inscriptions have been found on pottery west of the
Rhine (e.g. RIG I) and only occasionally on coins (no examples in RIG IV. 6—21; 27—
35; otherwise a few Vogelkopfstatere). The alphabets in common use there were Greek
and Latin, occasionally the Celtiberian (e.g. RIG IV. 319-321 no.200) and Lugano
scripts (as to stone incisions in Cisalpine Gaulish: RIG II. 3-8; Lambert 1994: 79).
The archaeologist I. Kappel, in her still authoritative treatment of the matter
(1969: 104—122), identified seven different types of marks, M. Cizmár and J. Meduna
(Meduna 1980: 100 f; both 1985; Cizmár 1995) increased this number to twelve. As
their arrangements - as previous investigation in general — strongly focus on the
‘cock-tread’ or ‘tree’ symbol and tend to neglect differences among the others, I sug¬
gest the classification in figure 5 with 94 types at present.
The order of signs is preliminary. It seemed most convenient to arrange the marks
according to their shape as e.g. in lists of cuneiform signs. There are two main
groups, signs written with straight lines and those with curved lines. Rectilinear
marks are further subdivided into horizontals, verticals, diagonals and two more com¬
plex subgroups, wave or zigzag lines and ‘closed’ figures as triangles and quadran¬
gles. Curved lines are subdivided into curves, including spirals, and circles or ovals.
Dotted marks have been attached to the end of the latter but could be regarded a
group of its own. ‘St Andrew’s cross’ (+ D5) has been classed with the diagonals
because it is often difficult to distinguish between X and ~K Within these groups, the
arrangement is usually from simple to more complex forms, yet it is not always possi¬
ble to propose a coherent order. Moreover, some signs which are rare or of limited
usage have been placed after more common ones even if they are simpler in form, e.g.
the diagonal ‘slash’ D14 (/) which is known from Norican coarse grey ware only.
The structured order of signs allows the inclusion or addition of marks not yet dis¬
covered, published or collected where they typologically belong, e.g. A Clb: shape
C (verticals, Ψ-shape) — type 1 (‘angular’ types) - subtype b (A). The table tries, as
far as possible, not to anticipate an interpretation of the marks in any direction
(%· 2).
Fig. 2: Typical graphite vessels with comb decoration
(from Stradonice, left and Manching, middle and right, scale 1:5, from Kappel 1969: left 98 fig.
43. 4; middle pi. 23. 650; right pi. 24. 669)
Fig. 3: Sample of marks on graphite vessels
(scale c. 1:5, source: Kappel 1969, from Manching if not otherwise stated: first row pi. 21. 619
and 630; pi. 23. 659; p. 102 fig. 48. 6 (Devin okr. Bratislava); second row pi. 26. 699; pi. 37.
879; p. 93 fig. 14 (Hallstatt Dammwiese); third row p. 102 fig. 48. 7 (Devin okr. Bratislava); p.
102 fig. 48. 3 (Vienna Leopoldsberg); pi. 26. 700
Fig. 4: A tripod vessel and a lid of a different pot made of coarse grey clay
and a sample of marks on native grey ware
(from Magdalensberg, scale c. 1:4, from Zabehlicky-Scheffenegger 1997: left top 130 fig. 3.1;
left bottom 128 fig. 1.5; right 131.12-14; 17-19; 22-24)
A number of theories have been proposed over the years to explain the function
and meaning of vessel marks. None of them have proved to be totally convincing.
Seemingly, it cannot be expected to solve all the problems connected with these
markings at once, and probably it must be conceded that no single hypothesis can
explain the phenomenon as a whole. So the arguments must be weighed carefully.
Marks as ornaments
Sometimes motifs similar to vessel marks appear as ornaments on the smooth
wheel-thrown La Tène ware, such as V or X (Cl, D2; see Pingel 1971: pi. 97.1515;
pi. 35.439; 440). As early as 1940, M. Hell (1940: 123) saw a connection between
ornaments on the sides of receptacles from the Fritzen-Sanzeno culture with the
‘cock-tread’ sign Y. Another possible example is the circle with a dot in the centre
in Postoupky (Cizmár & Meduna 1985; H3 O) which is smaller in size and, above all,
stamped, not carved into the clay. But to most researchers in the field, it seems
strange to assume decoration of utility pots on the exterior side of the bottom where
they remained invisible when in use. Moreover, real ornaments usually look rather
different (cf. e.g. Pingel 1971: pis. 93—97; Uenze 1993) and often occur as multiple
motifs in a string on the vessel side. So the explanation of the markings as pure orna¬
ments has long been discarded (as already Kappel 1969: 121). Hell’s comparison was
Straight lines
Al* -, A2* —, A3* —, ·-·
Verticals (with extensions except group C and + shapes)
/T, Y
B10 1
Bll Ej,e,
B13° -HB14° B
B15° :i(?)
Verticals with two diagonal extensions (E-types, ‘cock-tread7‘tree’ symbols)
Ψ, A, A
y, r
Cil y (?)
C13° Y
C15 Y(?)
Diagonals, crosses and multiple lines (Λ-, X-, and *-types)
Λ, Λ, V
X, x,x
+, +
D6* /\
X, »<, X
D13° Â
D14° /
D15 //,//
Wave and zigzag lines
H, « (?)
ΛΛΛ, ---. ΛΛΛ" E4*
E6* v</
Wnv/ * IvwÂl
E10° ·£(?)
Eir Ht
Triangles and quadrangles
N, X
F3* □
F5* D>
F6° me
F7° LBT (?)
Curved lines and dots
Curves and spirals
X, X (?), I
Circles and ovals (single and double)
Ο, ο, ·ό·
©, ©
Θ, ©
Θ, Φ, ©, 0
H9* Φ, Φ (?)
H10* © (?)
Dots and dotted figures
*) signs occur on metal objects only
°) signs occur on coarse grey ware only
Fig. 5 : Preliminary sign list of vessel marks
L, U
o, $ (?)
perhaps inspired by the romantic theory on the origin of the runes from ‘pre-runic
symbols’ (vorrunische Sinnzeichen ) of the bronze age which was current in the thirties
(e.g. Krause 1937: 425); cf. also p. 80.
Marks as labels of salt cellars
At an early stage of the investigation, the discovery of marks on sites like Hallstatt and Salzburg suggested a connection with salt production and salt trading along
the rivers. So the explanation of marked vessels as salt cellars has been put forward
(Hell 1940: 121; Franz 1942: 45 1). Later on, Hell (1952: 84) modified his opinion in
assuming graphite vessels as a kind of accompanying goods of salt traders — which
seems to be a watered-down version in the light of new findings from distant places. It
is very unlikely, however, that salt was transported in comb decorated, large open
graphite bowls even if receptacles were not mere ‘packing material’ in antiquity. The
vessels could of course be employed as kitchen utensils by salt traders but such a use
would not be restricted to them and would also be possible for other people. But, most
important, the distribution of the different types of clay does not support this assump¬
tion (Waldhauser 1992; Herrmann 2000). And the dispersal map of marks known
today does not show any more a significant concentration along the major river tradeways or a proximity to centres of salt production. So there is little support for this
hypothesis now.
Marks as a cipher system
Time and again the marks have been seen as number symbols. Two possibilities
can be imagined: ciphers to indicate measurements (Sinnhuber 1949: 105; on cipher
forms, see Cagnat 1914: 29—34, cf. also Radke 1967: 430) or output figures of a pot¬
ter’s workshop (as e.g. in La Graufesenque, Marichal 1988: 113-225). But nobody
has ever been able to detect a correspondence between the capacity of vessels and
incised markings; one may compare e.g. the cock-treads of type lc listed by Kappel
(1969: pi. 22 f) to get an idea of the diversity of pots marked with that symbol. In
addition, no cipher system needs sixty or more different symbols, and even the lowest
previously suggested number, twelve (Cizmár & Meduna 1985; Cizmár 1995), pre¬
supposes either very high numbers in a cipher system or a duodecimal or higher
radix notation system. In both cases one wonders why there are hardly any vessels
with two, three or more marks, corresponding to ciphers as ‘XIII’ or ‘13’, for instance.
Thus, the interpretation as ciphers rests on a weak foundation.
Though a vigesimal system is probably original in Celtic languages (Russell 1995: 90)
and a number system with 20 ciphers therefore imaginable, there is no evidence of a radix
notation system before the Arabs who took it over from India in the early Middle Ages. The
Brittonic languages preserve the vigesimal system up to this day (except ‘Living Welsh’), in
Goidelic it exists beside the decimal system since the Old Irish period (abandoned in stan¬
dard Modern Irish). There are traces in French (except in Belgium and Switzerland) sug¬
gesting its presence in Gaulish, but unfortunately no evidence from Continental Celtic is
available. Outside Celtic, we find a vigesimal system in Europe only in Basque and Cauca¬
Marks as potters’ brands
Most interpreters favour the theory that vessel marks are brands (e.g. Kappel
1969: 120-122; Pescheck 1976: 371; Lasch 1999: 48). And indeed, the fact that the
signs are scratched into the unbaken clay seems to support this idea. Some scholars
even tried to identify the potters’ workshops, such as wave lines (El— 2) in Milovice,
Moravia (Cizmár & Meduna 1985). The combination of ornaments and marks on
ceramics of the Fritzen-Sanzeno culture has already been used by Hell (1940: 121 f)
as an argument in favour of this idea. Hell did not think of individual imaginative
symbols, but of single letters derived from north Italic alphabets which were used as
brands. In spite of these contributions, however, some doubt should be cast on the
use of marks as potters’ brands. Firstly, most of the signs seem to be scattered over
the whole territory and probably distinguish local variants, e.g. Y as a western and T
as an eastern form of the same sign (Cizmár & Meduna 1985: 81). This is more likely
an argument for local differences of a script than for brands. Secondly, the theory
cannot explain why types of valuable ceramics, as e.g. the fine wheel-thrown ware or
the La Tène painted ware, have never been marked with signs on the bottom, contrary
to ordinary cooking pots and other simple wares. Usually one would assume to find
brands especially on expensive or richly decorated vessels. Thirdly, the same marks
have occasionally been incised after burning (post cocturam ) in imported Samian
ware (cf. Hell 1949) which points to a different intention. Fourthly, a recently
published item from Berching-Pollanten (Lasch 1999: 49) shows the variant H5c of
the θ-shaped sign on the exterior and a new variant D8, roughly resembling a star, on
the interior side of the same pot. This suggests two sharing or subsequent users rather
than co-operating potters. And a last point can be made: As Jiri Waldhauser (1992)
has shown in a detailed study, graphite vessels usually concentrate around the
natural deposits of the raw material, and scientific analyses confirm the local ceramic
production (Herrmann 2000: 21). From the findings, Waldhauser suggests trade with
finished products in a zone of c. 50 km, with clods of graphite clay up to c. 100 km
around the deposits, and only sporadic exchange of pots at a longer distance (1992:
380 1). But the dispersal of vessel marks does not follow this pattern. The oppida
Manching, Hallstatt, Trísov, Staré Hradisko and Magdalensberg show virtually all
signs known to this day but each in comparatively small quantities. Even smaller
settlements which have yielded the same marks, are often far from each other (on sign
El-2, see below). Fragments, however, which do not allow a clear identification of
the sign and have thus usually been attributed to ‘type 1’ (i.e. the whole C-class in the
present typology), must not be accounted for.
In Milovice, okr. Breclav, 13 examples of types El—2 have been unearthed in a potter’s
workshop, a pit house, a pit nearby, and a kiln (Cizmár 1994) where graphite clay pots have
been manufactured on a local scale. The vessels may have been transported to near-by
settlements, where types El—2 have shown up, such as Dmholec and Klentnice, situated in
the vicinity of Milovice, and probably to the oppidum Staré Hradisko, c. 75 km north to it.
M. Cizmár and J. Meduna (1985) infer this from typical Milovice forms on these sites. But
whereas c. 75 km roughly agree with Waldhauser’s model, it seems unlikely that recepta-
des have been exported to so distant a place as Karlstein, Lkr. Berchtesgaden, c. 400 km
from its suggested place of origin. A broken sherd with ends of two wave lines has been
found there (Kappel 1969: pi. 31,9), possibly made of “lower Austrian or Moravian gra¬
phite clay” (ibid.: 188; cf. p. 82). On the other hand, the raw material used in Milovice
came from southern Bohemia (Cesky Krumlov) and testifies to a trading distance of c. 175
km to the ‘eastern group’ of the graphite vessel production in Moravia (Cizmár 1994: 88 f;
cf. Horakova-Jansova 1955).
As one would expect casual concentrations of the same marks in a particular
area, the actual findings do not support the theory of potters’ brands. This does not
preclude, however, that some of the marks have been intended in this way.
Marks as symbols of religious or magical character
After dismissing the last hypothesis, Hell (1935: 183) explained the marks as
religious symbols or magical ‘salutary signs’ (,Heilszeichen ), perhaps again influenced
by contemporary interpretations of the runes. An assumption like that cannot easily
be disproved and indeed often serves in prehistory as a last resort. But certain instan¬
ces of La Tène religious symbols, such as spirals, triskels, swastikas, crescents, Shooks, dotted circles or eyes, plants, animals and the like are almost absent from the
sign list of vessel marks. Only star and spiral motifs could be referred to:
(D4) and
Φ (G5), the circle(s) with a dot in the centre (Θ, © H3— 4) as ‘eyes’, perhaps, and zig¬
zag lines as a ‘water’ symbol (El— 5). Θ (H5) as a ‘wheel’ is problematic since wheels
in iconography use to have 6—8 spokes (on symbols in general, see e.g. Green 1989;
Uenze 1993). So there are only weak indications to this hypothesis (cf. Kappel 1969:
122). Moreover, in recent years it has been emphasized that ancient scripts were not
restricted to a single usage, such as economic / administrative or religious / magical,
but that they were open to different applications at the same time (e.g. Nedoma 1998:
27; 46 f on the early function of the runes). Thus there is little in favour of this idea in
general, but a religious background of single marks cannot be ruled out.
This is all the more remarkable as it is a common tradition in north Italic and
runic inscriptions that they sometimes consist both of true letters and non-letter sym¬
bols. This is usually mentioned incidentally (probably because epigraphists are not
very interested in symbols), but R. Hachmann (1993: 328 f; 373-393; 415-420 fig.
8—13) dedicated a considerable part of his study to this subject. He explains the sym¬
bols as property marks and ‘signs endowed with power’ (jkraftgeladene Zeichen ,
p. 383), and in collecting and discussing the material he goes as far as the tamgas of
the north Pontic area. He also mentions Heilszeichen of late La Tène origin, such as
circles with dots, crescents, S-shaped signs, triskels, groups of three dots etc. on
Damascened spearheads, belt buckles from Holstein, and the sheet metals of two
chariots from Dejbjerg (Klindt-Jensen 1949). So there seems indeed to be a connec¬
tion between letters and symbols, and these observations do not simplify the inter¬
pretation of the vessel marks. But as mentioned above, only few marks are similar to
symbols and can probably be identified in most cases.
Marks as letters
As almost all ideas on the meaning of La Tène marks go back to M. Hell, this is
also true concerning the connection of marks with writing (1952: 85). As mentioned
above, his opinion is that isolated letters have been taken from north Italic alphabets
and used as brands of potters’ workshops. I. Kappel obviously follows him but does
not agree that all signs can be regarded as letters, e.g. the circles and wave lines. But
a comparison of the complete sign inventory with Alpine alphabets easily reveals that
the marks are not isolated borrowings since almost all signs on late La Tène recepta¬
cles (and similar ones on coins) have counterparts in north Italic writing systems (see
below). It will be seen (p. 105) that the idea of an adoption of select letters is absurd.
Nevertheless, in view of the large dispersal area and hundreds of individually deter¬
mined records it cannot be excluded that there are some artificial symbols among the
them.The letter theory must not preclude the possibility of a different origin of some
The idea that vessel marks represent a complete alphabet is corroborated by the
following facts: (1) A great number of sign types is present on each of the larger sites.
Often the same types are attested at different locations (bold) or at least similar types
Manching: Bl; C2, C3, C4 , C5; Gl, G3; H5, H8
Hallstatt: C2, C3, C4, C5, C7, C8 , C9; G6
Staré Hradisko: Bl, B2, B3, B7; CIO; D2, D3; E2; Fl; H4
(2) The dispersal area of signs with high attestation covers the same vast territory
(a selection of sites is given only):
Bl: Manching, Katsch, Magdalensberg, Staré Hradisko
B7: (Nußdorf), Magdalensberg, Staré Hradisko; (Norican inscriptions)
CI: (Hallstatt), Linz, Katsch, Magdalensberg, Stradonice, Trisov
C2: Manching, Kelheim, Hallstatt, Magdalensberg, Linz, Vienna, Staré Hradisko
C3: Manching, Kelheim, Hallstatt, Linz, Stradonice, Trisov
D2: Berching, Manching, Hallstatt, Magdalensberg, Staré Hradisko, Vienna,
Gl— 3: Manching, Kelheim, Devin (Bratislava)
Hl-4: Berching, Kelheim, Magdalensberg, Staré Hradisko, Nejdek, Mistrin
H5: Stuttgart, Manching, Pfaffenberg, Vienna
It will become apparent that almost all marks found up to now have counterparts
in the writing systems of the Alps and northern Italy. Seemingly, it was not isolated
letters which were taken over by the eastern La Tène population, as previously
accepted, but a complete alphabet.
In a formal analysis, vessel marks can be compared with similar letter forms in
contemporary alphabets. So far nothing can be inferred concerning the phonetic
values of the signs. They are treated as purely graphic symbols and transliterated, if
at all, with double pointed brackets «... » which give the value in the original script
for comparative reasons. In the following presentation, I must confine myself to a
summary in the form of a table and the discussion of a few examples.
The sign of the Ψ type (the 4 cock-tread’ or 4tree’ symbol)
The most common mark is the so-called Hahnentritt symbol because one type
closely resembles a hen’s footprint. In Kappel’s analysis, most of the present C-types
are treated as one sign, but they better should be subdivided as suggested here. The
first of them (Cl) is a cock-tread with three lines of similar length without extension
of the middle line: Y. This as well as its inversion A\ resemble very closely Alpine
forms of (Etruscan) <a> or maybe <kh>. In Katsch in Styria and on Magdalensberg it
occurs in the latter form A in short inscriptions as well. The second subtype (C2) is a
cock-tread with a short line to the left and to the right and an extended line in the
middle: Y or T. Forms with straight side strokes prevail in the west, curved ones in
the east (Cizmár & Meduna 1985: 81). The sign has the orientation of Y as is obvious
from examples on coarse grey ware (fig. 4), from a double incision in Hallstatt (Hell
1949) and probably also from a ‘cursive’ form hastily jotted down (Hell 1922: fig.
1.1a; 1934: fig. 1.1). It is probably found in an inscription in Katsch again (Y Cll).
Its use in ERIPOYlOS, i.e. eribogios, <Y> as /k, g/ (RIG II/1-E3) from Gropello Cairoli, Pavia may also be compared. It is identical in shape with Etruscan and Greek
<kh> in western (‘red’) alphabets; Camunic, Lepontic, Magrè, Sanzeno and Katsch
have identical forms. A further subtype consists of various forms with a fourth stroke
crossing or inclined to the middle line: Y, Y (C4— 5). More or less symmetrical
variants X (C6), similar to Steinberg and runic <z>, were found in mound 17 in Rats¬
chendorf, Austria (end of 1st c. AD, Artner 1995), Lanzhot, okr. Breclav (Cizmár
1995: 115 fig. 6,10), in the shape X in a salt mine in Hallstatt (Barth 1984), and
again on coarse grey ware, slightly different in form: X. Kappel’s assumption of a
high percentage of cock-tread symbols on La Tène ceramics — she suggests a figure of
83% (1969: 110) — gives a wrong impression. The frequency of each of the proposed
types of the C-group is actually less than c. 22% (cf. below, p. 88).
The sign of the Y type
An interesting sign from Linz, Römerberg consists of a circle and a vertical line;
both elements are written in one stroke: Φ (G4). Probably the same sign, minted as a
circle and a vertical line 0 (once with an extended vertical, Φ, both H9*) and sur¬
rounded by two dots, occurs 253 times on rainbow cups in the Bochum hoard (Forrer
1910: 455; Roymans 2001: 102 f). Not much imagination is called for to see a sim¬
ilarity to Greek and Etruscan <ph>, i.e. Y and Alpine forms Φ, Φ etc., respectively.
In Yenetic, the same sign is <b> because no letter denoting the phoneme /b/ was in
use in the Etruscan model alphabet. G4 / H9* could be transliterated « ph » for the
The sign of the Θ type
A further example (H5—6) consists of a circle or a double circle and a cross
inscribed into them: Φ, ©; nos. H31 (O,®;®) may also be compared. They are
almost identical with Etruscan forms of <th> and Lepontic <t/d>. Type H3, however,
is not carved into the clay, but stamped into it and may therefore be regarded as an
ornament. But perhaps type F2 (Ê3) belongs here although the cross is added to a
square in this ease. The similarity or identity of the sign with various <th>-forms of
Etruscan and northern Italic needs no further comment (cf. }, Alpine Φ, ffl etc.).
Even the eastern Gaulish silver quinarii of the ΚΑΛΕΤΕΔΟ Y type 4 (see recently
Burkhardt 1998: 43) perhaps show the θ-shaped sign H5 as a substitute of Δ (in the
reduced form ΚΑΛΕΔ(Ο)Υ) which is present on the same spot under the horse of the
reverse side in the earlier types 2—3. It seems improbable that the Δ was omitted and a
mere ‘symbol’ added in its place as is generally assumed. If to be taken seriously, the sub¬
stitution points to the phonetic value of a voiced dental stop for Θ, similar to, but more
archaic than Venetic X (and runic M) <d>. Types 3—4 are dated to 100—80 BC {ibid. 62).
(r, r?)
λ, \
/\, // (var.); A
-H-; *
Select parallels
C, E, F, L, M, Ma, N, L, S, Y 1, ΕΠ
V (Este) Φ, Gu Ψ
« th I s », « b » (?) SI,MM,C1
« k », V < y >
<k>: C, E, L, M, S », N =1 (?); <y>: V >1
« r », « s + i » (?) <r>: El, F, Me k, Fa, La R, M
« e m h »,
<e>: N, Nu III, N ΤΠ, F M, L, M
E, S
« pi ip » (?)
C, L X C, V -J ;
<b>: C III, HH ; <h>: V III, B, Ψ
<m>: EW,M,S f\, C, E, L, V 'Ί
« 1 », « n »
<1>: C, E, L, No, S 4, V A
<n>: C, E, L, M, N, S, V Ί, M U
« t », No <s>,
<t>: L, M t, M, Ne T, N Ν'; <p>: S +
<s>: No T;
S <p>, CF <n>,
F <ï>
<n>: C 4, F + <ï>: F 4 (?)
« w»
S, Ye =1, Μ, Ν, No, V % E, S Λ; <f>: F 1
« 1 », « u »
<u>: C, M, N, S, Y Λ, C, V N, F ΙΊ
<1>: C A, C, Me Λ, F, Μ, N, V 1
Letter type
« i »
-, -,-M?)
1, :i; / (D14 ?)
Bl, B15°
« h »,
« a », « kh »
C, F H, H; F, S, V M; E, N, V B, M, N, S, V
N; <d>: V X
<a>: K Ψ, EI A, C, L, S A, C, L, S, V A,
<k(h)>: E, L, S, V Ψ, E, L, S, V VI/, \l/
Fig. 6 : Comparison of the La Tène marks with letters
y (?),γ
Y, Y, X; Φ
Y (a scratch?)
X, X, X; X
+, +
X,X (?),I
A, A
— r\ ; C
H, “ (?),
N, X; El
Φ, ©,
O, ©, ® (?),<§>
O», 3
Letter type
« kh »
Select parallels
<k(h)>: C, E, K, L, M, S Y
<g>: V Y
«y »
<u>: E Y, L Y, K, N Y
« z »,
F X, Y, St X, C Y, C *
V <d>, C <ph>
E Í, L, M *;
<d>: V X; <ph>: C %
«t »
<t>: C, L, M, N, S, V X, E, V t
(cf. « ks kh »),
<ks>: La X; <g>: F X
« c + c » (?)
<k>: C, E, No, Ca 3 (see D12)
« a»
C, L, S, V A, E, M A, V <K, E, L, S A,
C, N Λ, F, L 1
« c », « p » (?)
<k>: C, E, F, No >, C, E, No, Ca 3
<p>: C, E, L, No, S, Y 1, L, Ne, S Γ
V <j>, C <k>
<j>>: V II; <k>: C II, U, 1
« s / s », « m »
<s>: C, E, F, L, M, S, V 5, /, C, E, L *, *;
<s>: E, M, S, V M;
<m>: M, S A, C W, C, E, L, No, Y 'Ί
« s », « th »
<s>: C, F, L, Y (Casale) N; <d>: F N;
<t(h)>: EI, V E, E, [M], No ®, Θ, V Φ,
c ®, :·: (?)
« o », « th »
<o>: Ca, F, L, Ma, N, V $, L 9, 0, Gu, Μ, V
O, V; C, E, L, V O, E, L Θ
< t(h)>: E, Y Θ
« d » (?), « a »,
<d>: E 0, El, No Ά; <a>: M *1, E, L, S A;
«r »
<r>: L, M, Ne, S, V <, C, E, L, S, V 0, N 0
<p(h)>: E Φ, L, M, S, V Φ, M ❖,
« ph », « th »,
E Y M, S 9, t, F, S P
C <q>
<t>: V O; <q>: C Φ
« q » (?), « o » (?) <q>: E, Me ?; <o>: L f, Gu, M,V
Cl 1-12
C4-6, C14
D4 (?)
D2, D3 (?)
G7*, G8°
Problematic signs
B3 l\i may be a combination of letters, if a letter at all; B14° E may be an ornament; C3 Y,
Y perhaps are variants of C2, C4-6 or ligatures (?); C7-10 Y, Y, Y, Y may be variants of
C2 or of C4-6; C15° Y is perhaps a variant of Cl or C4 or a combination of C2 and Al (?);
D8 Y may be a variant of D9 or an ornament; D13° Â could be a double D1 or a variant of
C14° without vertical line; E5
may be a multiple incised E31 or consist of several
letters engraved over each other; E6 VY may consist of more than one letter; E10°
Ell0 ftt are combinations or ornaments; F3* □ may be a variant of F2 or of F4/H3; F7° lH
and F8° to] are rather doubtful; G2 X and G3 X may be variants or marked forms of Gl, or
may belong to types D4/D9; G5 <§> may be an ornament or possibly a variant of H2, H4 or the
like; G6 Θ may be a marked form of G4; G9°
maybe triple G7 (?); H8 Ò is an ornament
or for
a ligature
(?); H10*
15° ;©and
is doubtful;
16° ···· may
11 *be· punctuated
may be a ‘diacritical’
Al*-3* ormark
Bl. or a dot as used in E,
Fig. 6 : Comparison of the La Tène marks with letters (suite)
Abbreviations : C Camunic (‘Raetic’, Sondrio; Capodiponte), Ca Castaneda, Kt.
Graubünden, E Etruscan (north and south varieties are not distinguished), El ‘East
Italic’ (PID II), F Futhark (common Germanic runic script), Fa Faliscian, Gu Gurina
near Dellach, K Katsch, Styria, L Lepontic (Lugano script), La Latin, M Magrè
(‘Raetic’), Ma Manching, Ravaria, Me Messapie, N Norican (Magdalensberg, Carinthia), Ne Negau A— B, No Novilara, Nu Nußdorf (Chiemsee, Bavaria), Pt Ptuj, Slove¬
nia, R Ratschendorf, Styria, S Sanzeno (‘Raetic’, Bolzano), St Steinberg, Tyrol, V
Yenetic (Este, Padua, Vincenza not distinguished), Ye Verona-Platen, VN Valle di
The signs of the V and A type
Markings in the shape of V or Λ (Dl: Λ, V) and the similar type with two
unconnected diagonal strokes (D6:
D15 //) were known from Celtic coins
( Vogelkopfstatere ) since 1907 and R. Forrer proposed the reading <au> for the
group(s) ΑΛ, AZ, A/,
A// (1910: 480; Kappel 1976: 85; see below, p. 921).
The variant writings clearly show that the orientation of the sign is indifferent (V—Λ),
that both hastae need not be connected - at least on a hard surface —, and that signs
next to each other can be written in ligature, as the variant A/ indicates. In the
meantime, type D1 (Λ) came to light in 1977 in Oberthümau, Austria (Bauer & Mau¬
rer 1978), carved into the internal side of a graphite vessel. The coarse grey ware on
Magdalensberg provides examples of the same sign (Λ: Zabehlicky-Scheffenegger
1997: 131 no.18). From both sources a prevailing Λ-orientation of this sign becomes
clear. The mark is of a common northern Italic type, as in Camunic, Sanzeno, Venetic, mostly inverted and usually representing the phonetic value <u> though a deriva¬
tion from <1> or <a> cannot be ruled out for the first variant Λ.
The sign of the III type
An interesting case is a type found in Staré Hradisko (B7), consisting of three
equally long parallel strokes (III or ΠΪ). As one of the ends of all hastae is lost, it is not
certain whether there has been a fourth horizontal line connecting the three. But this
need not be the case since the sign III appears on the coarse grey ware on Magdalens¬
berg (Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198 no. 160; Zabe¬
hlicky-Scheffenegger 1997: 131 no. 29) and in Norican as a letter with and without a
connecting hasta. The phonetic value there is <e> as is clear from the female name
PETU (Petü , Egger 1968a: 37—40). The form ΠΤ is also known e.g. from the reverse of a
‘Celto-Illyrian’ coin type found in Austria (Forrer I. 238) with a Hercules head on the
obverse. Since /h/ does not exist as a phoneme in Gaulish, <e> could be the initial
letter of Herakles (or a prince’s name beginning with the same sound). So the vessel
mark is transliterated «e », but alternative renderings like « m » or « pi ip » (i.e. ΠΙ
or vice versa), or Venetic <h> cannot be excluded.
The Magdalensberg findings
The key to understanding the vessel marks is probably provided by the late La
Tène and early Roman findings on Magdalensberg near Klagenfurt. All possible com¬
binations of characteristics are present on this site: Inscriptions on import as well as
on native ware, personal names in Latin as well as in epichorian script, and incisions
made before as well as after burning. The following observations can be made (cf. fig.
6; Egger 1968: 274—276): Markings on import ware have always been incised after
burning and show a large variety of proper names. They are either given in full or
without grammatical endings, but they are rarely abbreviated to less than three let¬
ters. Markings on native ware, on the other hand, have been carved before as well as
after burning in about equal shares. As a rule, they consist of one or two letters only,
with the exception of Latin letters incised after burning. It seems clear that the pot¬
ters’ brands hypothesis cannot apply in all these cases. And all the different symbols
which were incised before burning taken together would indicate a high number of
almost contemporaneous potteries for Magdalensberg.
Marks incised after burning
. . . before burning
1. on import ware (Samian etc.) (* Norican names)
C. Don(nius)
2.1 on native ware, in Norican alphabet
A( )
T( )
E( )
V( )
E( )
P( )
2.2 on native ware, in Latin alphabet
M( )
A( )
Fig. 7: Letters on Magdalensberg vessels
T( )
U( )
]eik( )
0( )
The same applies for vessel marks in a purely native context, and additional
observations confirm Egger’s explanation. The general use of different letters on the
same site implies that the function of the markings is in all probability to distinguish
between owners of vessels. Seemingly there were no supra-regional centres of pro¬
duction of ordinary wares, and we know it was the raw material (graphite clay) that
was shipped to more distant places (Kappel 1969: 123 ff; Waldhauser 1992; Herr¬
mann 2000: 22 fig.9). Many workshops, therefore, used to work on a rather small
scale. So the customers could very well ask the potters in their neighbourhood to sign
the receptacles ordered by them.
This would be a quite natural explanation for the letters carved into the unbaken
clay. In 1st— 2nd century La Graufesenque, there are clear examples of potters’ signa-
I (J) L M N
Fig. 8: Frequency of initial letters of personal names from Roman Noricum
Colours in fig. 8: dark grey = initials of different names only; light grey = initials of all name
occurences; white = vocalic /i, u/ grouped together with consonantic /i (y), u (w)/, <j, v>.
tures (e.g. Litugenus fecit “Litugenus made [it]”) as well as notes of destination (Rosirum Adronico, “for Andronicus, son/slave of Rosirunus”, Marichal 1988: 229 f), both
types having been incised before burning.
One further argument in favour of the explanation of vessel marks as initials of
proper names can be put forward. It has been shown that the nomenclature of the eas¬
tern Alpine region remained conservative in early imperial times (cf. Dietz 2000:
1004; generally, Dobesch 1980: 436-441). A statistical analysis of native personal
names from Roman Noricum (Alföldy 1974: 232—238; 1977; Lochner 1989; Hainzmann 1996: 447159; cf. id. 1987), based on c. 1200 occurrences, reveals that the
most frequent ones begin with <a> 16%, <c> 15%, and <s> 12%. The material made
up of 58 epichorian names on lead labels from a vicus population in 2nd c. Kalsdorf,
Styria (Römer-Martijnse 1990; Hebert 1991: 285, fn. 18; Alföldy 1993: 18 f; 26)
shows a similar distribution, the three most frequent ones being again <a> at 19%,
<c> 16% and <s> 22%. A minor problem may be that a number of names with initial
<s> or <t> actually begins with a dental affricate [ts] (or the like) which is not repre¬
sented in Latin by a letter of its own (tau gallicum), but probably such a letter exists
in Norican (as in Lepontic). The vessel marks most often used, on the other hand,
have the shapes Y « kh » (C2/11/12) and Y, Y « kh a? » (C3), formerly understood
as variants of one and the same cock-tread sign. The frequency of the two groups is
21.5% (C2/11/12) and 15.1% (C3) resp., which is close to the figures given above for
the frequent Norican initials <c, a>. A total agreement cannot be expected because of
the differences between the Celtic and Roman periods. If /k C3 could be <a>, as A
Cl probably is, both forms would add up to 17.6%. Together with a fourth variant, Y
« z » (C4—6; D4;8— 9: 6.3%), all cock-treads amount to c. 43% of the discernable
5 %—
Fig. 9: Frequency of vessel marks
Latin marks on imported vessels also appear on a number of other Gallo-Roman
sites, cf. the material collected by L. Bakker and B. Galsterer-Kroll (1975) or
recently from Karden, Mosel (Nickel 1999: 141—145): A, XX, I, ]VL (Iulius ?), VA
(Valerius?) etc. (on graffiti, see also Scholz, RGA XII. 558—567).
The use of vessel marks
The discussion on p. 81 has already pointed out that markings are not likely to be
potters’ brands, neither as individual marks nor as letters. One argument was that
ordinary cooking pots were marked whereas valuable earthenware was not. A distinc¬
tion between cooking utensils of various owners does not make sense unless for peo¬
ple who assemble in the same place and at the same time and want to prevent a
confusion of pots. This could have been the case in large workshops, communal feed¬
ing, and long-term meetings of varying purposes, such as festivals, hearings and so
on. This was already proposed by R. Egger (1968: 276):
“Das Eingraben von Buchstaben ins noch ungebrannte Gefäß vollzog entweder der
Töpfer auf Bestellung des Käufers, oder möglich ist auch, daß der Käufer das Sigle seines
Namens in den noch weichen Ton eingrub. Für die Familie sind solche Signaturen über¬
flüssig, sinnvoll und praktisch aber werden sie, wenn ein Familienmitglied außer Haus an
einer Gruppenverpflegung teilnimmt und sein Geschirr mitzubringen hat. Das gilt zum
Beispiel für die Arbeit im großen Magdalensberg-Markt. . . . Gemeinsame Essen begleite¬
ten auch Feste aller Art, Gerichtstage und Konferenzen.”
Another possibility is suggested by M.J. Enright’s study (1996) who demonstrated
the increasing importance of warband organizations in late La Tène Europe. And
finally, large assemblies of clients and councils {concilia) on a local as well as a
supraregional level are well known from Caesar (e.g. Bellum Gallicum i. 19.4; i. 30.4;
ii. 4.4; iii. 18.7; v. 6.2; vii. 75.1). So the increasing use of property marks in the 1st c.
BC may be a concomitant phenomenon of the development of more complex social,
political or military structures.
Very rarely ‘longer’ inscriptions, i.e. consisting of more than one or two letters,
have been found on some sites in the eastern La Tène culture. Material from Magda¬
lensberg and some other Norican sites is not fully included here because it is pub¬
lished by R. Egger (1959; 1968a; 1968b; 1969), G. Moßler (1961; 1986) and B.
Hebert (1991) or will be studied in detail by E. Schindler-Kaudelka and S. Zabehlicky-Scheffenegger as already mentioned. The inscriptions of the important empo¬
rium Magdalensberg, whose name G. Dobesch (1997) thinks to have been Virunum,
sometimes present a confusion of Latin and Norican forms (e.g. Egger 1969: 385 fig.
A problem which appears with inscriptions is the determination of the direction of
writing. There are two methods to identify the orientation. Palaeographical devices
draw conclusions from ‘external’ formal evidence, e.g. the orientation of single signs,
the overlapping and condensation of signs at the end of the line, and linguistic devi¬
ces make use of ‘internal’ structural evidence, as phonological and morphological
features of words and names. In the material collected here, only the direction of
signs gives an ‘external’ indication. So the identification has to rely mainly on linguis¬
tic observations. Possible readings (see below) point out that the orientation was not
fixed but that sinistroverse writing prevails. The same can be observed in Lepontic
(PID II. 516; Solinas 1994: both directions are well attested in LT Dl), Camunic and
Magrè. The Venetic script has also a predilection for the sinistroverse (c. 70%:
Lejeune 1974: 180 f), which was even more prominent in Etruscan. But both systems
rarely display dextroverse and boustrophedon inscriptions as well. The oldest runic
Futhark also shows sinistroverse, dextroverse and boustrophedon writing directions
(Krause & Jankuhn 1966: 2 f). In runology, this feature is regarded as one of the
‘archaic traits’ of the script (Antonsen 1996: 7). And going further back in the pedi¬
gree of alphabetical writing, archaic Greek stone and vase inscriptions (e.g. Board-
man 1974: 218—220) have the same ambiguity though they prefer the direction from
left to right.
In the previous chapters, the single letters attested could only be described as ‘La
Tène characters’ because nothing was certain about the language for which they were
designed. The following inscriptions give some direct allusions to the linguistic
milieu in parts of the eastern La Tène culture. From ancient history and onomastic
studies, of course, a Celtic language can be expected, but personal names like
Kenno-(?) or Atepo-/ Ateko-(?) provide some positive evidence. The language written
in ‘La Tène characters’ is likely to be Continental Celtic, probably a kind of ‘East
The inscriptions collected here are likely to be written in the same script as the
one on La Tène vessels but, of course, there are still many uncertainties. The exam¬
ples can serve only as a repertoire of possible testimonies and for some of them, a dif¬
ferent explanation may be offered in the future. The records have been treated in
three steps. Firstly, orthographic comparison with adjacent alphabets, in particular
Norican evidence, leads to a range of possible phonetic values (cf. fig. 6); secondly,
the phonemic and lexical structure of possible names and words helps to restrict these
formal possibilities and perhaps to identify some sound values; thirdly, the statistical
analysis on p. 87-88 probably gives some clues to the frequency of signs and phone¬
tic values. This does not mean, however, that a cryptographic approach is possible,
identifying most frequent signs with most frequent values. This goal would afford
many more records. As must be admitted, there is the slight possibility that the
inscriptions cannot be read at all because they are a kind of pseudo-texts according
to M. Lejeune’s definition, just imitating letters. ‘Texts’ of this sort have been found in
Este, Sesto Calende and other sites, but they usually occur on the margins of writing
provinces. It would be strange to assume such a vast area of pseudo-texts as that of La
Tène marks.
[1] In the 1880s a bronze vessel inscribed with four letters was excavated in a
grave in Zliv u Libáne (Schulz 1885/86: 80; Sakai 1965: 159) which is no longer
extant now. The signs are rendered by V. Schulz as vX V Λ (possibly upside down :
Y ΛΧ/*). The direction of writing cannot be determined with certainty, and the let¬
ters themselves are ambiguous. But it is important to note that the inscription shows
the X-sign (D2) to be a true letter.
1 u (a 1 ?) — u/v — t/d k/g — m s θ (e)
m s θ (e) — t/d k/g — u/v — 1 u (a 1 ?)
There are several choices for the rendering of a personal name, if this was really
intended. If <1> could be established for the first letter from the right (for <tuk/g/(t)/d> or
<auk/g> remain unexplicable, cf. KGPN 142), among others, the initials <luk/g...-> or
<lut...-> come readily to mind. If the ‘X’ can be explained as <t> (see below, no. 3), the
component lüt(u)-‘rage’ would be possible (DAG 237; 244; KGPN 233 f; cf. GPN 218, but
Old Irish hith, mentioned there, is probably not related: J. Uhlich, pers. comm.), as e.g. in
Lütumäros ‘great of rage’ (cf. DLG 178), known from Roman Noricum and Germania Supe¬
rior. Other possibilities would result e.g. in lugu-‘Lugus’, lukt-‘load, burden’ (KGPN 233,
luxtos DLG 179) or a name derived from Indo-European *leukl luk-‘shine’ (IEW 687-
690). Various alternatives, including <stul/t->, <skul/t->, <et/dul/t->, <ek/gul/t-> do not
result in attested proper names, but due to the high number of possible readings, nothing
can be established with certainty.
[2] An early Roman burial site in Katsch in Styria produced a single marking Ë3
(F2, perhaps a variant of H5; Ehrenreich 1994: 17; 34 pl.5 no. 1), hitherto unknown
on vessels, as well as a short inscription of four letters on the exterior side of a vase
with comb decoration: ΧΨΚΥ . To my knowledge, no Latin letters have been found
on the site. The short incision is remarkable as it shows two cock-tread variants in
context with other letters.
<— a k/g z/ts — k iy (?) — al k/g z/ts — 1 k/g
— > 1 k/g — a k/g z/ts — k (i)y (?) — al k/g z/ts
As the second and fourth letters (from right to left) seem to have consonantic values
only, the third must be vocalic because no combinations -kKt-, -kKK-, -kZt-, -kZK-, -yKt-, yKK-, -yZt-, -yZK-are to be expected in languages around the Alps. Thus, type Cl of the
cock-tread symbol is most probably to be read as <a>. For all four letters, possible rea¬
dings would be, among others, <ziyat/k, kiyat/k, k/tay(i)z, k/tay(i)k>, K being <y, yi, iy> as
in Venetic, and if the sign '/ could be a variant of <a>, <akat/k, ayat/k, k/taka, k/taya> as
well. Norican proper names include e.g. Caio/*Caia, matching a reading <kaya>, Caixu
<kayk>, Ciantullus <k(i)ya(n)t>), Diacu? <z?iyak>, Taio/*Taia <taya> (Alföldy 1974:
[3] In the Empress Christina tunnel (Christinastollen), a salt mine in Hallstatt,
four planks of mine timber with thin incisions have been found in 1980 (Barth 1984:
70 f) which have a 14C-date of 220 ± 80 BC (calibrated 380-170 BC with 68.2% pro¬
bability, OxCal 2000; most likely c. 3rd c. BC, see now Stöllner 1999: 48 f). The
planks show three inscriptions, one of them recurring twice:
(a) I AX A
item 6 S [A]; no. 7 S:
Bl, Cl, C6, Cl
item 7
E5 (= E2/3/4?)
item 8
Dl, FI, D2, B9 (2x)
Although the explanation of the marks as evidence of writing has been denied by
the excavator, the shape and use of the signs show a remarkable resemblence to the
eastern La Tène script and thus may be taken into account here.
[3a] The vertical stroke stands a bit apart from the rest but may nevertheless be
regarded as the letter <i>.
Reading from right to left, <atsai> may be suggested, with two slightly different forms
of <a> in no. 7 S (the first one is not preserved in no. 6 S) and a star-like tau gallicum (ts),
without vertical strokes to the left and right (as in M and t>KI). It could be understood as a
noun at?ä ‘rib; side’ in the dative-locative case of the feminine ä-declension (Lambert
1994: 55): alai ‘on the side’ (<
The noun derives from Common Celtic *ast-‘rib,
side, lath’, Welsh *as (plural eis, assau, i.e. an original o/u-stem), Welsh, Middle Cornish
asen, Old Irish asna (LEIA A-94 f; OIL 55; GPC I. 219 f), perhaps Celtiberian azas
(gen.sg., tessera K.0.1, if z < *dj,, dd, R de Bernardo Stempel, pers.comm.; cf. Meid 2000:
18—24), and ultimately from Indo-European host-hp/n-‘bone’ (cf. Hittite hastai-, Sans¬
krit ásthi-, Avestic ast-, Greek όστεον, Latin os, Venetic dat.pl. ostinobos, see KWAi I. 150
f; Elboume 2000: 17 1). The mention of a point of destination on prefabricated slices of
timber would be quite imaginable.
[3b] may be a single letter <mlsle> (?) shown off with double and triple carvings
or several letters engraved over each other (cf. examples from Lepontic: Solinas
1994: 360 no. 94 (1); 365 no. 104; J.F. Eska, pers. comm.).
[3c] repeats the same order of four signs, the last of them being reflected. Not long
ago, M has been identified as a vessel mark in Malé Hradisko (Cizmár 1995: 111 fig.
Preliminary remark: All script abbreviations are as in the legend of fig. 6.—
t?ln?ly?ls? — tl k(h)/g — t(h)/d z — u/v 1
u/v 1 — t(h)/d z — t k(h)/g — t?ln?ly?ls?
The only plausible readings seem to be with vowels in the first and last positions (V:
unknown vowel) <Vktu, Vgdu>, or <utkV, udgV>, possibly also <Vd/tzu, uzt/dV>, less
probably Cud/t-g/kn-, ud/t-ks-, ud/t-g/ky->.
In the light of [3a], the suggestion may be made to read <itsu>, with a variant (?)/f of
<i> (cf. V II, >1, » <y>, F fc) <y>, 1 <i ?>), the X-shaped sign representing <t> as in V, L, C,
B, S, M, N, and the Λ-sign as <u> as in V, C, S, M, N, F. The “segno a farfalla”, M is likely
to be a kind of <s>, here written <s>, which occurs in E, L, S, N, (in F as <d>) and in a
comparable form again in V (?! s2). A ‘Celto-Illyrian’ coin type from Ribnjacka (Bjelovar,
Croatia: J. Winkler in Forrer II. 64; Pink 1974: 36 = 1939: 54 f) with a tournament rider
shows a rather similar form in its sinistroverse legend $·Ι3ΙΦΧΙ9ΧΙ <sastieni.s> (with E,
V punctuation?) which represents the name of the Macedonian leader, Sosthenes (279/78
BC). it-su would be the locative plural in -su of a Common Celtic noun *id- ‘foot’, with
generalized dehnstufe from Indo-European *ped- (Sanskrit pad-, Greek ττοΟς", ποδό$,
Latin pes, pedis, Gothic fôtus, EWAi II, 77). The original meaning of *id-su (assimilated
Itsu) ‘on the feet’ developed into ‘below, under’, as in Old Irish is, Welsh is (DIL 413; GPC
I. 2031; cf. Old Church Slavonic podu ‘under’). The reading presents an additional pro¬
blem as the word is not written with tau gallicum like the first one in [3a] though in (late)
Gaulish it is also used at morpheme boundaries, as in bue Θ < *bhuet-s(e) (3rd singular pre¬
sent subjunctive of ‘to be’ + -s(e) [Larzac], see Lambert 1994: 171). The plank marked with
this word served as a footboard in the tunnel, so an explanation like the above would be
quite suitable. But due to many uncertainties this can only be taken as a very vague first
Reference may be made to the fact that the carpenter belongs to the class of privi¬
leged people (Kelly 1988: 61 f as to Old Irish law) and that his very name, Common
Celtic *sa(q>)eros (Irish sáer, Welsh saer, from Irish?), derives from Indo-European
*sapero-, {LEIA S-6; cf. Latin sapio ) ‘the skilled, expert one’. Therefore it would not
be totally unexpected if he was able to read and write.
[4] Gold staters with a bird head ( Vogelkopfstater) of the Mardorf type from a large
area in Germany, Austria and Luxemburg display various forms of an incision
consisting of two letters: ΑΛ, AF, A/, A\/, A// (Blanchet 1905: 462; Forrer 1910;
DAG 1181; Kappel 1976: 88-90 lists 24 occurences, see above 3.4; Wigg & Riederer
1998), a form A Λ is known from Irsching, Bavaria (Forrer 1910: 479 f). In the light
of the variant renderings, Forrer’s reading <au> is convincing, but one must account
for <ua> as well.
[5] Triquetrum rainbow cups of the Bochum type, named after a hoard found in
1907, first published by Buchenau in 1908 (see Forrer 1910: 455) and most recently
studied by N. Roymans (2001), display part of an alphabet in the form of extra marks
(Beizeichen ) on the reverse. Royman’s new typology lists the following (2001: 102 f.
fig. 4, his types in minuscules): A (r) «a», (D (n; s) «ph», Φ (k) «th», — (d) «?», — (i)
«?», © (q) «thlo», I or — (f; p) «i», o (b; c; g; m; p; q) «o», + (o) «t», V (1) «u»; there
is one clear letter combination: — o (p) «ioloi» (Forrer <io>), and a doubtful second,
0· (n). Buchenau and Forrer (I. 454 f) also listed
Θ and the variant Φ which were
omitted by Roymans because of their doubtful identification. Except the ‘A’, and of
course the almost ubiquitous Ί, O, V’, the signs have little in common with the Latin
alphabet, but again resemble the La Tène (and Norican) scripts rather closely. It
could very well be that the (Etruscan) aspirate letters Φ, Θ represent voiced stops
<b> and <d> as in Venetic. — and — could be taken as renderings of « E » and « F »
[6] On Boian ‘staters of the older gold coinage’ from a Manching hoard (Ziegaus
2000; Leicht & Ziegaus 2000) of the 2nd— 1st c. BC, one or two letter combinations
have been found. One type of staters (ibid.: 380 f. no. 6, quoting NemeskalovaJiroudkova 1998: no. 6—18) shows three (?) letters AO\Ä/.
<— (ligature ?) I m (?) I s/s — o — a I v
—» a v — o — s/s I m (?) (ligature ?)
If a word or name was intended, the second letter could most plausibly be taken as
<o>, an alternative <th> being rather improbable. Then, the A-like letter possibly repre¬
sents <v> rather than <a>. And if <vo-> can be regarded as the initials, it could be the first
letters of personal names with vo-‘under’, e.g. Vosegos ( GPN 288 f; KGPN 299 f; DAG 83;
237) or Volto-(‘hair’? KGPN 300; DLG 275) etc. On the other hand, the letter on the right
side resembles the <aulua> combination briefly discussed in [4a]. So again, little can be
said with certainty.
A combination of markings Χλλλ is attested several times in the same hoard but
could be an ornament or a non-alphabetic marking by the mintmaster.
[7] Nike gold staters from the beginning of the Bohemian coinage (c. 250 BC, see
Ziegaus 1997) sometimes show legends which are not likely to be ‘barbarisations’ of
the original Greek words BAEIAEQC ΑΛΕΞΑΝΔΡΟΥ, “King Alexander” (genitive) but
substitutes in a native script as e.g. TEULO (De la Tour, pi. 40 no. 9474).
As the letters <e, o> seem clear, and LL are likely to be variants of <n> or maybe <1>,
<?enno/(?ello)> would be good choices. The direction of reading is perhaps determined by
the orientation of <e, n/l>. From left to right, Kenno-(KGPN 170: Ceno-, GPN 175 ff; DAG
140; 228), from *ken-‘spring, issue’, keno- ‘long?’ (DLG 97) or from archaic *k“ennom
‘head, top’ and ‘chief in the light of Welsh pen-, un-ben (cf. the place name element -cenna
as in Sumelocenna, de Bernardo Stempel, RCA XVI. 412) or maybe Kello-(KGPN 170; cf.
GPN 332 f; IEW 546, from *kel- ‘strike, hew’ (kellos DLG 96) or *kelh- ‘ragen,
hoch/heben]’; Venetic Ca 5: ke.l.lo.s, Fogolari & Prosdocimi 1988: 309) would be possible,
thus providing the value <k> for the Ψ-shaped cock-tread. Name statistics on p. 87-88
already gave a hint at one of the values <a, k, s> for this common vessel mark. So <k>
would be a perfect fit. Less likely is a sinistroverse <onne?> which could be connected
with onno-/a (GPN 370 f; DAG 54), Welsh onn-en ‘ash tree’, Irish onn ‘pine tree’ (doubtful
onno ‘flumen’, Endlicher’s Glossary).
[8] A second Nike stater reads □<IIIXt> (De la Tour, pi. 40, no. 9475) with
obviously similar signs as e.g. on the Hallstatt mine timber.
One has to be cautious, however, for the inscription may be a degenerated part of
the Greek legend (2nd line) with some deviations (3rd 1.).
< — d/t r I ο (a) — d/t z k(h)/g — ij/ji? e pi/ip — g/k p 1 n — t(h)/d o
—> t(h)/d o — g/k I p 1 1 1 n — ij/ji? I e I pi/ip — d/t I z k(h)/g — d/t I r I ο I (a)
Provided that the legend is a personal or ethnic name, like most other lexemes on
coins, I see four possibilities, all reading from right to left: (1) <ateko>, as e.g. in Atecotti
(GPN 144); (2) <adeko>, cf. A(n)deco (GPN 138) or initial letters of A(n)de-ko(m-bogios),
attested several times, also in Etruscan (KGPN 128; GPN 153); (3) <atekt> for atekto-,
ateyto-(KGPN 137; GPN 265; DLG 50) which is less probable because of the different let¬
ters <t>; (4) <atepo>, i.e. ate-(prefix ‘very, over, re-’, GPN 142 f) + epo-‘horse’ (GPN 197
f), known in a number of compounds (GPN 198 f; KGPN 137 f). If one of these explanations
holds true, it would confirm the letters X <t> and III <e>, show a rectangular variant of <o>,
□ (with a dot, also present in Venetic: Θ, ) and would add rotated or simplified forms of
<a>, D> (cf. >). One letter cannot be clearly identified: < may be a rotated gamma <g>,
perhaps <k> or even <p> (cf. f, Λ). A dextroverse alternative would be a name composed
with <o(r)keto?>, Ore/gelo-etc. (KGPN 252 f; GPN 239 f; DAG 177; DLG 206) ‘slaying’, as
the famous Orgetorix·, <k/g> would be an expected value of <3, but two different shapes of
<o> are not likely.
[9] Among the inscriptions listed here, some uncertain examples and possible
ligatures may be reckoned as well. The existence of ligatures is almost certain from
the variants in (4a); cf. also examples from Celtiberian (MLH IV. 613 f; 651 f. nos.
K.1.13— 15; K.1.19; K.5.6 and K.5.8).
[9a] A probable instance is a combination in Vienna Leopoldsberg (Kappel 1969:
102 fig. 48.3) of a wave line with two fragmentary diagonal strokes which may derive
from the V-type Dl, X-shaped D2 or λ-shaped D7:
or C'y. This may also be a
hint at an alphabetical value of the wave line which has usuálly five archs and has
previously been taken as an ornament (Kappel 1969: 120 as to her type 7).
[9b] The 1986 excavation in Berching-Pollanten, Lkr. Neumarkt i.d. Oberpfalz,
revealed two signs engraved over each other, the circle sign O (HI) and a hitherto
unknown variant Ï of Gl with intersecting curves: ώ (M. Leicht, pers. comm.; not in
Lasch 1999); Gl has been carved onto HI, both into the moist clay, so the possibility
can be ruled out that Gl is of secondary origin.
[9c] A ball-shaped dish from Tauberbischofsheim has incised lines on its side
which could be a combination of signs (D2, Bl, ?): ψ\ (Hoppe 1982: 161 fig.42.2).
But it is not clear if it is an ornament, a symbol (?) or a graffito.
[10] An inscription already mentioned shows some resemblence to the Norican
and La Tène scripts and is, therefore, discussed here in short. In 1995, some unusual
finds were made at Nußdorf, Lkr. Traunstein, near Chiemsee in Bavaria (Ziegaus &
Rix 1998; incisions on bone from Farchant near Garmisch-Partenkirchen mentioned
on p. 297 fn. 30, seem to be less comparable, Lang & Schultz 1995: fig. 42). Most
remarkable were a small silver coin minted in Marseille prior to 470 BC (tritartemo rion, Furtwängler’s “monnaies du type d’Auriol” Qe/Qf, 1978), a Hallstatt Fußzier¬
fibel of south-west German or east French provenance with a date of c. 510-470 BC,
and a small silver ring (0 18.8—20.2 mm, weight 0.83 g) with an inscription of
unknown date. The three pieces have been found within 30 m on a plateau. There is
no evidence that all three were deposited at the same time. Other fibulas and frag¬
ments found on the same site were of late Hallstatt (D3, c. 500—480 BC) and middle
La Tène (c. 3rd— 2nd c. BC) date. The inscription was interpreted by H. Rix who sug¬
gested a ‘Raetie’ legend in the Sondrio alphabet, a local variety of Camunic :
íLivVi/ναΠΐΡλι 1 1
My attempt differs from Rix’s in several points. Neither the direction of writing nor the
starting point of the inscription are clear. I retain the arrangement of Ziegaus & Rix (1998:
292c-e) and comment the latter’s reading. Rix omits the first sign on the right side III (cf.
B7) which occurs probably twice on the ring. Its phonetic value may be <e> as in Norican
(see p. 101). The next sign is X (cf. D2), possibly <t> as suggested. The sign group (?)o·'-3
is explained as <ip> neither of which I would confirm. The hook e3 was passed over but
reminds of <p> or could be part of angular shapes of <h, th, s> or a ligature containing
other letters. The sign a seems clear and could represent <r> as well as a broad quadran¬
gular <o> (as to broad forms in Norican: Moßler 1961: 559), but also forms of <h> or even
<th> are possible. The two shapes v (cf. Die) and y (cf. D7) are likely to represent <u>
and <1>, <a> or another shape of <u> (‘y’) respectively, less likely is Rix’s <X>. The next
combination -<vVA is difficult to analyze, the first might be <n>, as assumed by Rix, or
<m>, or even <k> (?).
could be <a> as Rix suggested, and -Z4 may be <w> (or this sign
may be <a>). +— as <i> seems an unusual reading, and <t> would be equally possible. 4 is
dubious (<p s η y>?). ''s could again be a Norican shape of <e>, and III is quite diffi¬
cult. The line ends with III (cf. B7) and V/NA which need not be a letter but if it is, an <m>
would be most likely:
<-e (?) — t (?) — (p.. h th s ??) — r ο (h th) — u (?) — 1 a (u X) — (η I m
z? k?) — ? — ? — (a?) — a I w ? — t (?) — ? — e (?) — z (??) — e (?) — m (?)
-> m (?)— e (?) — z (??) — e (?) — ? — t (?) — a w I ? — (a?) — ? — ? — (η m I
z? k?) — 1 I a (u x) — u (?) — r ο (h th) — ?(?) — t (?) — e (?)
Thus, I would not be so confident concerning a plausible reading of the ring
inscription and an identification of the language as ‘Raetie’ which seems little more
than derived from the supposed Sondrio origin of the script. Since several letters do
not exist in this alphabet (III, the zeta and khi forms) but in Norican and La Tène ves¬
sel marks, the alternative of a native origin of the ring inscription in middle La Tène
(3rd-2nd c. BC) may be suggested. This being accepted, the inscription could possi¬
bly be Celtic and, if turned upside down and read from right to left, follow a formula
like me ... (NN) ... (verb)-te “NN made (or sim.) me (?)”, with <me> as the accusa¬
tive of the 1st singular personal pronoun as attested in Voltino, to-me-d(e)-ek-laï
Obalda natina (Eska & Weiss 1996; to-me-(i)d- Schmidt 1980: 190, against Meid
1989: 20—26), and a verb in 3rd singular preterite with the ending ~(t)e, cf. Lepontic
karite in Vergiate (Motta 2000: 196; Meid 1999: 14 f), Gaulish tosokote in Vercelli
(Lambert 1994: 64; RIG II/l, *E2), or old perfect forms as dede etc.; less likely would
heme ... (NN) ... “I (am) (product of NN)” for which the use of immi Ί am’ could be
The material collected here, meagre as it is, has important implications on the
significance of writing in late Continental Celtic societies. One must bear in mind
that the major part of the evidence is lost in the archaeological record since it was
made up of temporary notes on perishable materials such as wax tablets or wood
(Kruta 2000: 406; Markey 2001: 79; cf. Eck 1998). Thus our knowledge is extremely
fragmentary, a fact also strengthened by L. Pauli (1992: 620). In the similar case of
literacy in Etruria and Latium in the archaic age, T. Cornell (1991: 24—32) empha¬
sizes that the role of writing in these societies has largely been underestimated.
The dispersal of writing across the Alps
Our picture of the early dissemination of writing to the north and east of the Alps
has considerably changed over the last decades. Today we know that Celtic tribes
adopted and adapted alphabets as early as the 7th c. BC, as is clear now from a graf¬
fito on a vessel from a Hallstatt settlement at Montmorot, Jura of c. 600 BC (<pris>:
Verger 1998), and perhaps from two letters on a golden bowl in a prince’s tomb at
Apremont (Haute-Saône, <VX>: Mohen et al. 1987: 75-94) of about the same time.
Another early instance of writing north of the Alps is a terracotta tablet from Ramsautal, Dürrnberg near Hallein, with seven indistinct (Greek?) letters on the reverse,
dating from the 4th— 3rd c. BC (Zeller 1989; Christian according to Eichner). Graffiti
from Manching date to the 2nd or early 1st c. BC, as the well known BO IOC and
ΖΗΘ inscriptions (Krämer 1982; 1996: 368; Eichner 1989: 18—20) and a recently
published sherd with Latin letters TAR[ ] (Schubert 2002). They are probably Greek;
the letter <B> of the first is distinctively Latin or Greek and occurs also in Novilara,
Ptuj in Slovenia (Eichner 1994) and runic Futhark (cf. on Greek letters e.g. Duval
1989; Lambert 1992; Lejeune 1989). We may also mention the glass bead from Miinsingen-Rain, Kt. Bern of c. 300 BC with an inscription (Vitali & Kaenel 2000: 121;
<samoritos> (?): Gambari & Kaenel 2001) and the incision of the KOPIC IOC
sword from Port-Nidau, Kt. Bern (Wyss 1954), as well as Greek and Latin coin
legends. An interesting document is the zinc tablet from Bem-Thormebodewald
(Fellmann 1991; 1999). It dates, however, from the Roman period, which is
comparatively rich in records in the Gaulish language (cf. Lambert 2002).
The most important early evidence is made up of Lepontic inscriptions, the first of
which is from Castelletto Ticino of c. 570/50 BC (<Xosioiso>: Gambari & Colonna
1988; Prosdocimi 1991: 142-144; in general see Leponti 2000). Beside the Lepontic
or Lugano script, different local alphabets were in use, as Camunic in Valcamonica
and the Norican alphabet(s) on the Gurina near Dellach, Deutsch-Landsberg, Styria,
and Magdalensberg (Egger 1959; Moßler 1961; Hebert 1991). The last three sites
have yielded a considerable amount of vessel marks and short inscriptions, which
were studied for the first time by G. Moßler (1961; 1986; Eichner 1994: 138). They
are now collected and further examined by S. Zabehlicky-Scheffenegger (1997; 1998;
in prep.). It would be little surprise if the vessel marks testify to a further example of
a native script. Although we have but a few short inscriptions, this disadvantage is
balanced by the fact that there are hundreds of single marks which cover a wide area
of dispersal.
Archaeology provides additional evidence as writing implements have been found
in several oppida , such as bone pencils or styli (stiti, στύλοι) from Bem-Engehalbinsel, Manching, the Závist, Staré Hradisko and the Hradist near Stradonice (Jacobi
1974; Drda & Rybová 1995: 190; Kruta 2001: 594; cf. Lenerz-de Wilde 1987); the
last site yielded the bronze frame of a wax tablet (Jacobi 1974: 172). And not surpris¬
ingly, in the emporium on Magdalensberg styli, writing tablets, ink pots and seal
cases (Piccottini 1989: 15) have been found. So there is both direct and indirect evi¬
dence of the practice of writing and a large part of the eastern La Tène culture can no
longer be regarded a white spot on the map of the history of writing.
Moreover, the Venetic alphabet was in use as early as the 5th c. BC in Carinthia,
Styria and Slovenia (Lejeune 1974: 285-309; Istenic 1985; Eichner 1994: 138; Gobi
1989: 13; 27; Marinetti 1998), e.g. on the Gurina near Dellach (Jablonka 1993;
2001). But contrary to M. Lejeune (1991; 1992), there are no Venetic inscriptions at
Szentlorinc in Hungary (Eichner 1994: 137 f; Prosdocimi 1995: 77-79). ‘Raetie’
inscriptions are known from Steinberg in Tyrol and other sites in the northern Alps
from the 5th-4th c. BC on (Ziegaus & Rix 1998; Schumacher 1998; 1993; 1992; cf.
Mancini 1998; on the language: Rix 1997a). Contacts between Celtic tribes and their
neighbours have recently been treated with growing interest, particularly rela¬
tionships between Celts and ‘Raetians’ (Uslar 1996; Räter 1992), as well as Celts and
Etruscans (Vitali & Kaenel 2000; Dobesch 1992; Etrusker 1992).
At a first glance, some Gaulish letter forms, especially those from the Rhone val¬
ley (Allen & Nash 1980: 116 f; RIG IV. 11—15), seem to resemble La Tène markings,
as A <a> (= Cl), k <1> (~A D7), Θ <o> (= H3), V, /\ <u> (= D1/D6) and perhaps
others, but most of them occur as regular variants in Latin epigraphy as well (see e.g.
Cagnat 1914: 11—24; DAG 1372, except for the ‘dotted o’). The sign 0 (= H5) can be
explained as a ‘monogramme’ (RIG IV. 15) or a letter denoting tau gallicum. The
Lugano script in north Italy and a local script on a fibula from Castaneda (Kt.
Graubünden, Switzerland), are also essentially different, but there seems to be a con¬
nection between these alphabets and vessel marks which perhaps gives a hint at a
common descent. In the Lepontic Golasecca culture, marks on the bottom of native
vessels already existed in the 6th c. BC, as e.g. X and >K in Sesto Calende (Binaghi &
Rocca 1999: 440 f; Solinas 1994: 366 f, nos. 106; 109). Incised markings repeatedly
occur on late La Tène ceramics from Giubiasco, Ornavasso-San Bernardo and
Solduno (Stöckli 1975: 100-104), but they are obviously written in the Lugano script
and Lepontic language; as to Lepontic texts, see M. Lejeune (1971) and P. Solinas
(1994; 1999 [inscr. near Verona]). Incisions in San Bernardo and Giubiasco, for
example, are sometimes rather similar to La Tène marks north and east of the Alps,
such as (h (San Bernardo: Graue 1974: pi. 18,2), Ik (ibid., pi. 18,3),
(pi. 21,2), W
(30,1; 39,3), l (44,3; IV 46,7; 54,1; 59,9), © (67,4; 74,7; 74,6; see also Uslar 1991:
149). As the kind of relation between these markings and the La Tène script is not
clear, the problem must be postponed for the moment.
Based on classical authors, it is usually taken for granted that Celtic scripts west and
and north of the Alps are Greek in character. But this commonplace should be treated with
caution since Romans took old Italic and Alpine scripts for ‘Greek’ whatever their real ori¬
gin was (Radke 1967: 402). Thus the Old Latin inscriptions in Rome (Dionysius of Halic¬
arnassus, Antiquities 4.26; Tacitus, Annals xi.14.3), the Oscian script (Varro apud Servius
Auctus 8.564) as well as the Helvetian records mentioned by Caesar (Bellum Gallicum
i.29.1) and the unknown script north of the Alps in Tacitus (Germania 3) were all regarded
as ‘Greek’ which need not have been the case. On the other hand, R. Gobi’s opinion (1973:
24; 87) that the name BOIO on Norican coins is Venetic cannot be maintained because <Φ>
is used instead of <B> in Venetic proper, as e.g. in Es 28 <cpo.i.iio.s> Boios (but <B> is
present in the Ptuj alphabet).
Nevertheless, not each kind of marking can be accepted as evidence of a script.
There has been a controversy concerning notch woods found on the Kelchalm near
Kitzbühel, Tyrol, from the llth-8th c. BC (Altheim & Trautmann 1939; 1942; Pittioni
1942; 1950; Krause 1943: 241—246). But nowadays most scholars are inclined to
take them as property marks or aids for payroll accounting without any connection to
writing (Preuschen & Pittioni 1954: 87 f; cf. Goldenberg 2002). So-called Beigla for
fixing shares of Alpine pastures, as known from folklore studies, may be compared
(Liechtensteinisches LM 1975: 64 f). Similar markings on planks discovered in
Ockenhausen, Lkr. Leer (Fansa & Schneider 1993; see RGA XII. 559; 8th c. BC) may
be marks for construction. Furthermore, cross-shaped signs and vertical strokes on
stones and bronze utensils from the urnfield and Hallstatt periods (Hallstatt A/B,
13th—8th c., and C/D, 8th-5th c. BC, resp.; Asen 1903; Mayer 1976) and sporadically
on ceramics of these periods (Kappel 1969: 119 fn. 232; Willvonseder 1932: 275)
cannot be regarded as evidence of writing and occur as isolated markings from the
Bronze to the Middle Ages (e.g. Mayer 1976: 370; Cech 1989: 199 f).
From the foregoing it apparently need not be the case that there was one single
alphabet in so large a territory as the dispersal area of vessel marks, which was the
home of different tribes such as the Vindelici, Boii, ‘Norici’ (cf. Dobesch 1980: 274—
278), and others. Perhaps there were several local variants or even different writing
systems, each of them covering only part of the area.
A tentative phonetic interpretation of the La Tene alphabet(s)
Based on the comparison of the La Tène and other Alpine scripts, only a few sug¬
gestions could be made (see fig. 6), fully depending on the shape of the signs. A num¬
ber of c. 30 signs clearly points to a kind of alphabetic script. But as no clearly
intelligible inscriptions have come to light yet, the identification of values is difficult.
Thus, a few observations must do for the moment.
Generally speaking, all phonemes of a language should be represented in an
alphabetic script to avoid ambiguities. But minor exceptions are usually acceptable
as e.g. the absence of phonemic vowels in Egyptian and some Semitic scripts. If the
use of the alphabet is restricted, even a considerable number of exceptions seems
tolerable, as e.g. the lacking distinction between voiced, voiceless and aspirate stops
(e.g. <Pa> for /ba, pa, pha/) or lacking notation of syllable-closing nasals in Linear B,
resulting in numerous ambiguities like <Pa-Te> for /pántes/ ‘all’, possibly also
*/pater/ ‘father’. Allophones are usually not written with characters of their own. But
as a matter of fact, there are examples, as the letters <q> (‘ng’) in Indian Devanägari
and runic Futhark. In the environs of the Alps, the northern Etruscan alphabet was
adopted which did not represent voiced stops and the vowel <o> because the respective
sounds were absent from its phonemic inventory. Though appropriate letters were
needed for Indo-European languages, the deficiency was not repaired in Lepontic
with the exception of <o> re-introduced from (Etruscan) abecedaria. (Lateran, a dis¬
tinction in the dorsals <k> and <g> was attempted but did not gain acceptance.) In
Venetic, the situation was different. Besides the ‘resurrection’ of <o>, letters denoting
aspirate /ph/, /kh/ and affricate /z/ (?) were used to represent the voiced stops /b/, /g/, /
d/ in Venetic. As in Etruscan, but contrary to Latin, /u/ and /w/ were distinguished,
and a similar contrast between /i/ and /y/ gave rise to a new letter shape <y> (Lejeune
1991: 789). In Camunic, original <b, g, d> were retained. Thus there is evidence of
different treatments of the model alphabet which led to ‘weakly’ and ‘strongly’ devel¬
oped derivatives.
Later Gaulish shows some remarkable evidence of phonetic accuracy. The Latin alpha¬
bet did not distinguish between vocalic /i, u/ and consonantic /y, w/ whereas in Gallo-Latin
inscriptions consonantic /y/ is often represented as iota longum in the combination -iyV~,
as e.g. in <Rll0S> (Corthals 1999: 101; cf. Marichal 1988: 60-65; RIG ILI. 144). Contrary
to this usage, Latin i longum was used to denote the old diphthong ei and monophthongized
i, respectively (Cagnat 1914: 17). In the early imperial period, however, it became a mere
graphic variant of normal i and thus could not serve as a paragon for Gaulish. A similar dif¬
ference between /u/ <ou> and /w/ <u> in Gallo-Greek has been indicated but is difficult to
reconstruct. The distinctive allophone [x] was represented by <χ> or <x>, even in GalloLatin where a homographie <x> [ks] existed; /ts/ was represented by tau gallicum , barred
d, s and the like (Eska 1998; the phonemic status seems clear from minimal pairs as e.g.
dong’ : dir-‘star’ : *Ur-‘land’). On Celtiberian <z> for affricate [ts] or voiced [z], less
probably [9], see Jordán (1998: 25-28) and Meid’s valuable overview (2000: 18-24). The
runic script shows a remarkable ‘perfect fit’ as well, cf. e.g. R.L.M. Derolez (1998b).
The inventory of La Tène marks (see fig. 5) points to a number of c. 22 letters
which would rather indicate an alphabet of the Camunic or Venetic type than a limi¬
ted one like Lepontic. The number of ‘East Gaulish’ phonemes, on the other hand,
can be estimated to about 30 (or c. 21 without long vowels and diphthongs) under the
condition that the language does not differ much from the Gaulish in Gaul or what we
know of Galatian (Freeman 2001; Schmidt 1994). The Gaulish phonemic inventory
includes the following (see Lambert 1994; 40-47; as to [æ], [i]: McCone 1996: 55—
57; Uhlich 1999; Indo-European equivalents cannot be given here):
/.../ indicates phonemes, i.e. sounds which can distinguish between meanings, as /k/
and /g/ in English ‘cap’ against ‘gap’, [...] allophones, i.e. sound variants which cannot
distinguish between meanings, e.g. the different ¿-sounds in ‘top’ and ‘tree’).
/a J
voiceless /ts/
voiceless /p/
/er/ /i:/
/au/ /eu/ /ou/
m ?)
(loll sporadic or late)
(/oi/ is rare)
(lïl is uncertain)
(/kw/ is rare)
(/gw/ is uncertain)
<h> in Latin renderings is merely orthographic, cf. Haedui : Aedui
Differences between voiceless aspirated (?) fortis [ph, th, kh] and non-aspirated
lenis [p, t, k] on one side, and voiced plosive and labial nasal fortis [b, d, g, m] and
fricative lenis [β, δ, γ, μ] on the other side are probably already in existence, but not
yet phonemic (see McCone 1996: 84—88 on continental Celtic lenition; de Bernardo
Stempel, in print, on deo Xuban in Aquitania with <x> [y]).
Very little can be said with certainty about the values of the La Tène letters. Most
identifications result from the shape of the signs and some hints from the interpreta¬
tion of the few short inscriptions.
The vowel letters may be known in total, at least in some of their forms. It is not to
be expected that short and long vowels are distinguished.
A; A; V, A (DIO*; Dll; Cl) clearly seem to represent <a>. A (Dll) and A
(Cl) occur as variants in coin legends [4] <aulua>, A (Dll) perhaps in [6], and V,
A (Cl) is also present in inscriptions [2] and [3a], <atsai(?)>, with a plausible expla¬
nation. A (Dll) is similar to Norican <a> as well.
E, III (Bll*; B7) are probably variants of <e>; both appear in inscriptions, [7]
<kenno(?)> and [8] <atep/ko(?)>. Norican <e> resembles III (B7) closely.
I, — (Bl; A2*) correspond to the common shape of <i>, also found in Norican.
Both occur in inscriptions [5] <ioloi> (?) and again in [3a], see above sub <a>.
O, £; O; © (F4*; Hl; H2) are likely to be shapes of <o>, the first variant appear¬
ing in inscription [7] <kenno(?)>. Norican <o> has shapes like £ (F4*).
N (B12), and perhaps Λ; /\, // (Dl; D6*, D15) as well, could be variants of <u>
because in [3c], the sign is found between the word boundary and a combination of
the ‘farfalla’, M (FI), and X (D2) which are most likely to represent consonants. Nori¬
can (and Venetic) possess the same inverted <u>.
Semivowels and sonants are far more difficult to identify.
1 (BIO) resembles the shape of digamma, <w> which appears in Norican in a
slightly different form Λ, but the same script has perhaps a variant of <k> which
looks very much alike (). The sign did not yet occur in inscriptions.
L, U (B8*) represent probably <n>, less likely <1>. Northern Italic parallels,
Norican, and the reading of [7] <kenno> (?) seem to plead for <n> but in the light of a
possible <kello>, <1> cannot be ruled out.
Ά (B6) reminds of <r> but it is attested only in early Roman context in Katsch
(2nd c. AD) and therefore may be a loan from Latin. Norican, Venetic and others sug¬
gest a D-like shape (perhaps attested in F5* ?).
Other consonants cannot be determined with certainty except in a few cases.
X, X, X (D2): From inscriptions [3c] <i'tsu(?)> and [8] <atep/ko(?)>, perhaps [2]
if it couldand
be Venetic.
<kiyat>, the assumption can be made that X (D2) represents <t> as in
Y, Ψ (C2), one variant of the ‘cock-tread’ sign, can possibly be read <k> as in [7]
<kenno>; it appears in ERIPOYlOS (Gropello Cairoli, cf. p. 82) with the value <K> = /
k, gI. The letter derives from Etruscan <kh> and could possibly provide an argument
for an origin of the La Tène script independent of the Venetic tradition (see p. 102)
which uses the same sign for <g> from the earliest records. The value <k> agrees
with the statistical analysis of personal names on p. 87-88, the high frequency of
marks corresponds to that of the name initial Ik-1.
X(D9), and most probably also Y, Y, X (C4; C5; C6) could possibly represent
tau gallicum, <ts>. This is suggested by the reading of inscription [3a] again, see
above sub <a>. A letter for this sound is not recorded in Norican and looks different
in Venetic, but similar to Camunic ‘tree’ forms of zeta.
The suggestion of further phonetic values would be mere guesswork, such as K (B5) as
<y> (?), see inscription [2]; < (D12*) as <g/k p I n> (?), see no. [7]. But the observation
that no B-and D-like letters are attested in respective quantities possibly implies that the
script is a “Phase Two or aev-Alphabet” after the elimination of ß, y, δ from the model
alphabet (c. 600 BC, cf. Markey 2001: 81-83). Instead of these, Φ (ph)-und Θ (th)-forms
may have served to represent <b> and <d>. Under this condition, Φ; Φ, Φ (G4; H9*) could
possibly render <b> and θ (H5) <d>. The last one appears as a variant of Greek Δ on
KALETEDU quinarii and is found besides + on triquetrum rainbow cups if it is allowed to
take these coins into account.
Derivation of the vessel marks from ‘northern Etruscan’
From the comparison of signs on p. 83-84, the assumption seems plausible that
the inventory of vessel marks is derived from a model alphabet of the Etruscan type.
The comparison with adjoining Alpine writing systems has shown a remarkable simi¬
larity in the letter shapes. And a collection of more than 1200 marks should in all
probability reveal a relatively complete alphabet. But in the sign list, clear testimo¬
nies of the common shapes of <b> (B4 ?), <d> (F5* ?) and perhaps <g> (D12*, G7*
?) are almost missing and, on the other hand, different letters <s> (El— 2 ? FI— 2 ?)
may exist. This is all exactly what one would expect of an Etruscan-based alphabet.
It is far more difficult to decide which one of the numerous northern Italic systems
based on Etruscan (or Yenetic: Rix 1997: 232) was the model alphabet. The most
exact parallels are probably provided by the Camunic, Venetic and Magrè scripts.
Especially interesting is Camunic (Morandi 1998; Tibiletti Bruno 1990; 1992) which
has the richest graphemic inventory and is attested from the 1st half of the 6th cen¬
tury on (cf. Markey 2001: 81-83). But only a few letter forms are distinctive. The
original theta shapes Ë3 (F2) and 0 (H5) were no longer in use in Lepontic and ‘Raetie’
but appear in Camunic and at least twice in Magrè and as a ‘dead letter’ in the Este
variety of Venetic. But as we know but few abecedaria of the smaller scripts, this
could be a spurious argument. Types Ύ, Ύ, X (C4-6) look very much like Camunic
<s>, but they are rather similar to zeta forms in Lugano, Bolzano and Magrè as well.
The ‘farfalla’ M (FI) is identical with the letter san in Lepontic and Camunic. On the
other hand, the sign
(B2) resembles Venetic i <h> very closely if B2 can be taken
seriously — it is attested only three times in Staré Hradisko. The F-shaped digamma
<w> is lacking in Camunic but seems to be present in Lepontic, Norican and La Tène
(1 B10). On the whole, it seems that in almost all cases an identical or very similar
sign can be found in the proposed northern Italic systems. The correspondences need
not be repeated here as they can easily be seen in fig. 6. Only in a few instances,
other Alpine forms are closer than the ones proposed (B6— 7). Furthermore, it seems
clear that vessel marks and Norican (Moßler 1961: 559; Egger 1968; 1968a) are
related but a full comparison cannot be made because the Norican inventory is still
rather fragmentary. Thus, a cross-check shows that no writing systems in northern
Italy but Camunic, Venetic and Magrè provide so many parallels to the La Tène script
though all of them are closely connected (cf. PID II. 505-520).
It is easily acceptable that speakers of Celtic languages could come into close
contact with literate people in the Alpine and North Italian areas. The Valcamonica
was invaded by Celtic tribes in the 4th c. BC and it is also known that there was a
Celtic presence in the Veneto (Marinetti 1998: 54—56; Gambacurta & Ruta Serafini
2001). Cross-cultural contacts are directly testified by a possibly Gaulish inscription
from Oderzo (*Od 7: Fogolari & Prosdocimi 1988: 303-307; Eska & Wallace 1999),
which mentions a person pompe-te(n)guaios, ‘(knowing) five languages’.
The age of the La Tene alphabet(s)
The earliest testimonies collected here go back to the 3rd century BC. These
include the Boian Nike gold staters of about the mid 3rd c. BC and probably the mine
timber from Hallstatt from c. 380—170 BC, most likely also from the 3rd c. BC. One of
the vessel marks from Frauenstein may be late middle La Tène (LT C2), i.e. 180—120
BC. Most of the evidence, however, comes from late La Tène and early Roman mate¬
rial, c. 120 BC — AD 150. Thus, archaeological dating provides a terminus ante quem
of about or shortly after 300 BC.
Palaeography corroborates such an early date. The Etruscan alphabet must have
been adopted independently in the Lugano area and the Veneto around 600 BC. At
that time, the Etruscan script was in sporadic use even in trans-Alpine Hallstatt set¬
tlements (Verger 1998; Mohen et al. 1987: 75-94). But since the majority of letter
shapes, including the ‘early’ theta forms EË3, Θ, appear in Camunic, Venetic (<D> in
abecedaria : Lejeune 1974: 32; 188; 195 [Es 23], cf. Padua »<) and in Magrè inscrip¬
tions in the 4th-2nd c. BC, the alphabet could have been adopted by Celtic literates
in the late 4th or early 3rd c. BC, i.e. not much prior to the first records. If arguments
in favour of an independent transfer from Etruscan could be found, in addition to the
probable use of Y for <k> (Venetic <g>), an adoption as early as the 5th c. BC would
be imaginable. But this question must be left open for the time being.
The origin of the runic script or Futhark is still disputed, and each theory proposed so
far has its pros and cons. A summary of the discussion is now presented by K. Diiwel
(2001: 175-181), and valuable methodological contributions are made by E.H. Antonsen
(1996; cf. 1982) and now by T. Markey (2001) but it is difficult to do justice to all scholars
in the field (cf. also Derolez 1998a; Griffiths 1999; Morris 1988; Odenstedt 1990 etc.). The
most favoured hypothesis today is also the oldest, L.F.A. Wimmer’s derivation from the
Latin alphabet (in his case, the Roman capitalis ) in 1874. He was followed, among others,
by the influential E. Moltke since 1951 (cf. Moltke 1985) who pleaded for an origin in Den¬
mark. An alternative, S. Bugge’s and 0. von Friesen’s recourse to the Greek minuscule of
the 3rd c. AD, first suggested in 1890, and the supposed authorship of the Goths, has large¬
ly been abandoned, particularly because of the probable date of the earliest runic finds in
the 2nd c. AD and the poor evidence connecting east Germanic tribes with the Futhark
(Hachmann 1993: 349-358; 1994/95). A new approach was made in 1928 when C.J.S.
Marstrander put forward his profound contribution of a north Italic or Alpine origin and a
possible Celtic or Marcomannic mediation (1928: 120-124; 185). His theory was soon sup¬
ported by M. Hammarström and was received with enthusiasm, especially on the continent
(e.g. Krause 1937; more cautiously 1966). For a long time the ‘northern Etruscan’ origin
was taken as the definite solution until the discussion eventually concentrated on the insol¬
uble problems of the period, place and circumstances of the transmission (see Rix 1992;
1997; cf. on the Negau B helmet: Nedoma 1995; Markey 2001a; Nedoma & Reichert
1998). The difficulties led to a reinforcement of Wimmer’s theory which has always been
the Scandinavian favourite choice (see Antonsen 1975; 1982; also A. Quak in Looijenga &
Quak 1996; B. Odenstedt and others in Bammesberger 1991; Hachmann 1993; Germanic
phonetic values of Latin letters used for writing names are discussed by Reichert 1987:
xix-xxviii). Whereas all theories mentioned so far started from a comparison of letter forms
(and phonetic values), today a new method is under consideration, the internal analysis of
the order of the alphabet (Seebold 1993; Griffiths 1999). But as the Futhark is not
completely standardized and the order is not known before the 5th c. AD (Kylver stone) and
the methods of ‘re-ordering’ are more or less arbitrary, this approach has its shortcuts as
well though the question of order is legitimate. The idea of a Celtic intermediary has never
been completely given up, and is brought back to the debate by E. Seebold (1991; 1993:
415), A. Griffiths (1999: 191) and B. Mees (1999). All three suggest a relationship of runes
and Ogam as well but this connection must be left out of account here.
The similarity between the La Tene alphabet(s) and the runes
The La Tène alphabet(s) as described in the last section show a remarkable simi¬
larity to the runic script as well (see fig. 10). From the point of view of letter forms
and values, and the archaic character of the script, the most convincing theory on the
origin of the runes is C. Marstrander’s derivation (1928) from northern Italic alpha¬
bets, recently reconsidered by E. Seebold (1991), H. Rix (1992; 1997) and B. Mees
(1999). No proposal but this one offers so many formal and structural parallels. In
particular, the archaic traits of the runes must be taken seriously because they were
part of the entire writing system adopted by early Germanic tribes (Antonsen 1996:
7). So these characteristics cannot be treated as marginal (as e.g. Quak 1996: 173).
They are a strong hint at an ancient Italic tradition, including such archaic features
as the variable writing direction (sinistroverse, dextroverse and boustrophedon ), the
rotation and inversion of forms, the use of word dividers, the differing height of let¬
ters, the lack of denotation of geminated consonants or nasals before tautosyllabic
consonants etc. A further argument in favour of the northern Italic hypothesis is the
use of letters of the Etruscan khi <kh> and theta <th> types to represent Germanic
voiced /g, §/ and /d, 9/. This is a common feature of Venetic and most Alpine scripts
which took it over from Etruscan. But it would be surprising if the ‘inventors’ of the
runes refuted Latin <G> and <D> in favour of <X> and double < Da > (?), in con¬
trast to <b> which they did use for /b, ß/.
The major problem with Marstrander’s theory is the eclecticism imputed to the
‘inventors’ of the Futhark. They are believed to have borrowed a few signs from
almost each of the Alpine scripts since some of the letters can be recognized as ‘typi¬
cally’ Lepontic, Camunic, ‘Raetic’, Yenetic and even Latin (see fig. 10). But this kind
of ‘letter collecting’ is absurd because the adoption of a script presupposes an
acquaintance with the phonemes of both the illiterate and the literate languages in
order to understand the function of an alphabet. Or, as E.H. Antonsen puts it,
“It has long been recognized in other quarters [i.e. outside runology] that alphabetic
writing systems are borrowed by illiterate cultures through individuals who have learned
the language of the literate culture and then the writing system of that culture, and only
then do they, or CAN they, attempt to adopt and adapt this foreign writing system to the
unwritten language.”
(Antonsen 1996: 7; similarly, Hachmann 1993: 339 f)
The adoption of (single) letters must be seen in this light too, such as the integra¬
tion of seven Demotic signs (e.g. < uj > fsf) into the Greek alphabet to establish the
Coptic script. This kind of taking over requires familiarity with the basic script
(Greek) in order to perceive its lack of accuracy in rendering some phonemes of the
foreign language (Coptic fsf).
The bilingualism implied excludes a casual adoption ‘in passing’, whether by the
Cimbri on the war-path or by Germanic soldiers in various services or by priests in
a— highly problematic — Germanic central sanctuary (as in Rix 1992). It also exclu¬
des the kind of ‘play with letters’ assumed by H. Williams (1996; 1997); e.g. he
thinks Latin <m> to be the origin of runic <e> only because of the similarity in
shape. Since Latin as well as Germanic possess the phonemes /e/ and /m/, a ‘switch’
or ‘mix-up’ of the letters would be hard to understand (likewise the change of <g> and
<y>, <k> and <p>, <o> and <q/q>). Actually no alphabet adoption is known which
works like that. T. Markey (2001: 87—89; 138 f, n.12) rightly emphasizes that “acqui¬
sition of literacy proceeds in pedagogical contexts” and that literacy “is de natura the
outcome of a diffusion contact process” (p. 138 f).
Furthermore, H. Williams’s explanation of the shape of runic <d> is not at all convinc¬
ing. An ‘invention’ (1996: 215) of runic <d> would have been unnecessary because neither
<rj> nor <i (?)> represent phonemes in Germanic — a fact obscured by Williams’s descrip¬
tion as “(perceived) phonemes” (1996: 216). So, there would have been a choice between
two Roman letters (Q, Y) to cover the phoneme thorn /0/. The hypothesis lacks an explana¬
tion why the notation of allophones should have been preferred to that of a phoneme.
Marstrander’s theory becomes attractive again as soon as an intermediary model
alphabet can be suggested and the gap in time and space of the transmission of that
alphabet can be closed. It seems reasonable to assume that the Norican or La Tène
script(s) under discussion here could possibly be such intermediary stages. It is clear
from the foregoing that the vessel and coin marks of the eastern La Tène culture
represent an alphabet most probably adopted (in several local versions?) from nor¬
thern Italy as early as c. the 4th-3rd c. BC. In this period, archaic traits are still exist¬
ing which appear in the runic as well as the La Tène scripts, such as the varying
direction of writing (sinistroverse p. 91f, dextroverse in no. [6]), orientation of signs
(V in [1], Λ in [3], λ and 'L within [3]), differing height of letters (in [7]), double and
triple incised letters, especially circles (H2; H4; H6), the cross sign (D2c) and zigzag
lines (El-2; E5), cf. ‘double runes’ like <s> in Bergakker (Bosnian & Looijenga
1996: 11), etc. E. Antonsen (1996: 7) emphasized the fact that the entire writing sys¬
tem is taken over “which includes intrinsic features as the direction of writing ...”.
Or, as T. Markey (2001: 90) puts it concisely: “Evidence from shapes is ancillary to
that from systems.”
Beside the evidence from the writing systems, the affinity of the La Tène letter
shapes (and values) to the runes is a strong argument as well. In most cases even our
restricted knowledge suggests Norican and La Tène letters as possible sources of the
runes. Particularly runes with a hitherto problematic pedigree find good matches,
such as <d (?), e, k2, p, r, w, y (?), z, θ>; k2 means the so-called ‘younger k’, Y, K, k
(Krause 1966: 113; Seebold 1991: 458 f; 488) which is, however, attested on the
Yimose woodplane of AD 100-300. The forms of <a, c (?), f/v, i, 1, m, o, s, t, u> are
also quite the same though similar letters can be found elsewhere. <b, g, ϊ, n, r)> are
problematic, partly because of contrary evidence (lack of <B>, <X> /t/, not /g/),
partly due to the restricted knowledge of the Celtic script(s).
C. Marstrander ultimately may have been right in assuming a Celtic intermediary
stage between the north Italic and runic alphabets. It is noteworthy that M. Hell saw
already a causal or mediating role of the Ψ-type vessel sign, which he, however,
regarded as a ‘tree of life’ and the origin of the Germanic mannaz rune (Hell 1940:
123). R. Pittioni (1942: 382 f) also thought positively about an adoption like this.
R. Egger, too, mentioned a possible connection between religious symbols in
Noricum and the runes (1953: 89 f). So the idea is not at all new but appears in a new
light due to recent discoveries and a new approach to their study.
Alternative considerations on the origin of the runes, particularly the theory of a
Latin derivation, suffer from a number of problems. E.H. Antonsen resumed that
“literally all of this evidence points away from the Romans on the Rhine as the
source of the Germanic futhark. We do not know from what exact source it sprang, but
what evidence there is points to an archaic Mediterranean script of Greek or Latin
origin” (1996: 11 f). He himself (1982; 1996) tried to identify the runes as an
“Ableger der großen archaisch-griechischen Schreibtradition” (which applies to the
Etruscan alphabet with its derivatives). C.B. Rüger (1998) who investigated Latin
literacy on the Germanic border of the Roman Empire was very reserved as to the
spread of the Latin script to the north: “Wir vermögen auch nach Öffnung des
weiten Feldes römischer Schriftkultur an der Nordwestgrenze nicht recht zu sehen,
wie von hier aus eine Anregung zum Runengebrauch ausgegangen sein sollte” (1998:
Possible ways of transmission
The eastern La Tène script or scripts were in use in the 3rd c. BC-lst c. AD for
restricted purposes on ceramics, coins and wood, perhaps on wax tablets, sometimes
also on other metal objects. From the unexpected high number of testimonies, it
seems that writing played a role in late eastern La Tène societies which is more
important than has hitherto been acknowledged. Widely distributed, as it is, the
and values
t>, Illerup Φ
La Tene marks,
(Φ, η ?) V <¡>
values and numbers
N <w>
Dl, D6
(V <t/d>)
k, \i,
Meldorf f?
<, r,r
Beuchte '/ (?)
L Illerup t
Μ, N
Λ, A
(r* ?), Y
N <r>
DU, Cl
Y <k> (?)
G7*, C2
X <t> (?)
(V <b>)
(V <h>)
N <n>
D2, Gl-3
G4; H9
Moltke 1985: 99 ff.
B9; B8*
Seebold 1991: 446
<i>(?) I
X, X, (X,X ?)
Φ; Φ,Φ
(>\< ? V Ψ)
{λ, Y ?; >, U ?),
N r
1, —
(K , N Ä ?),V «
(/Γ, Y ?)
(V <y>)
Bl, A2*
(< ?), N j
N <p>
X, Y
II, H, ΛΛΛ, vJ
X, +, N -Y, +
Vàdena t
Ill , N ÏÏT, III, II/, \
0#?), M /XV
A; (U, k ?), N r
(Ö ?), (E ?)
<ts> (?)
K Westeremden
t, M, Breza M
Í, i, $
(M <m>)
N <1>
(E <q=k>)
D7; B8*
M <s> (?),
® <d> (?)
O, O, N $
Fig. 10: Comparison of runes and La Tène marks
Fl, H5,
F4, H1
Μ, Π, (II),
Bergakker V
f\ Chamay k
o, □, Ο, Φ,
Vimose t
C6, D9
E 1 —4
<t>(?), N <t> D2, D5
N,®, Kl
Special comments
Seebold 1991: 445
Moltke 1985: 99 ff.;
Rix 1992
Seebold 1991: 445
Antonsen 1975: 9;
Mees 1997
Krause 1966: 26—29
Antonsen 1975: 2;
Seebold 1991: 469 f.
Seebold 1991: 444;
Rix 1992
Rix 1992
Seebold 1991: 447
Rix 1992;
Markey 2000
1996 & Looijenga
Krause 1966: 22
Seebold 1993: 415;
Griffiths 1999: 184 f
script could easily become known to neighbouring people in the north, especially in
Central Germany and Silesia, with a material culture like Jastorf, late Billendorf,
Kobyl, and Großromstedt. Cultural exchange between Jastorf and La Tène has
existed for a long time, and from late LT C2 to D2 (mid 2nd to 1st c. BC) there were
even two phases of Tatènization’ in the north, as has recently been investigated by
J. Brandt (2001: esp. 152). In the first century BC, the ‘Germanic’ material cultures
spread southward and interacted with the native La Tène in Hessia, Bavaria, Bohe¬
mia, and Moravia (Cizmár 1996; Waldhauser 1996; Venclová 1988; 1973) as well as
east of the Alps in Carinthia (Gleirscher 1996; 2001). The coexistence and eventual
fusion of old residents and new immigrants in late La Tène has been studied in
Altendorf, T ier. Bamberg (Pescheck 1977; 1978) and Aubstadt, Lkr. Rhön-Grabfeld
(Völling 1992; 1995) in northern Bavaria, and extensively in Radovesice and Horni
Lukovskÿ Potok in north-western Bohemia (Waldhauser 1992a; 1993; cf. Gosden
1984). The same overall picture is confirmed by findings in the Wetterau north of
Frankfurt where a mostly peaceful coexistence and an acculturation of La Tène and
Großromstedt populations is likely to have taken place. It seems that the lifestyles of
both groups complemented each other since the residents carried on agriculture and
craftmanship whereas the immigrants practiced animal husbandry. The latters also
seemed to appreciate the formers’ products such as salt and ceramics formed on a
potter’s wheel (Seidel 1999: 197-199). The presence of immigrant people at central
places like Dünsberg may, however, indicate their supremacy. In addition, a recent
research programme was concerned with the interactions between the Roman
Empire, ‘Celtic’ and ‘Germanic’ tribes in the central German highlands from c. 50 BC
to AD 300 in particular. “The finds indicate that [in the 1st c. BC — AD 1st c.] in Mainfranken, at least, the population consisted of a mixture of Celtic and Germanic ele¬
ments, who lived alongside each other, but in separate settlements” (A. Wigg 1999:
38; cf. Völling 1992) though only a few oppida survived into the latest La Tène period
(LT D2, c. 50—10 BC). — The term ‘Germanic’ in these statements is not to be taken in
the linguistic, but in the established archaeological sense of the word, based on the
Roman concept of Germani (Lund 1998). Even ‘Germanic’ pottery from Roman mili¬
tary sites in the Wetterau and the Lahn valley was attributed to native local inhabi¬
tants instead of ‘Germanic’ auxiliaries (A. Wigg 1999: 39). From the beginning of the
1st c. BC (LT Dl) on, finds of ‘east Germanic’ character are attested alongside late La
Tène material in Hessia, as e.g. in Mardorf (Meyer 1994; A. Wigg 1997: 59) and the
Dünsberg oppidum (Maier 1996/97). In Bohemia, a relationship to Central Germany
was established particularly for the late 2nd c. BC (LT C2— Dl: Salac 1992: 83;
Griinert 1992: 149 f). The Hradist near Stradonice continued into the middle of the
1st c. BC (LT D2, leading types: spoon-bow brooches, Almgren 65 forms) and there is
no evidence of a direct military attack. Instead, “the process of abandonment might
have lasted for a certain period” (Rybová & Drda 1994: 132).
Small wonder then that in Altendorf a mark of type C9 has been found on the nor¬
thern border of a ‘Germanic’ cemetery from the early Roman period (Pescheck 1976),
and an example of Cl has come to light in a grave of the Lübsow horizon in Prague 6Bubenec (Sakar 1965: 153 fig. 2,2). An urn cemetery of ‘Elbe Germanic’ character
near Wittenberge in Brandenburg has yielded a vessel with an impressed mark of the
Y -shaped C2 type (Schulz 1939: 152—155; fig. 188; cf. Franz 1942: 46, fn. 2). Some
ceramics on this site clearly imitate wheel-thrown wares and thus show that southern
imports were highly estimated. Furthermore, a bronze bucket of the last century BC
from Schlabitz (Tschiläsen) in Silesia, again with a Cl mark, may be quoted (Jahn
1919). The first collective study of Germanic settlements of the early Roman period
(Roman B2) in Moravia (Droberjar 1997) reveals that on six out of eleven selected
sites incised vessel marks of a similar kind as the La Tène marks were in use (Dro¬
berjar 1997: 41 fig. 26. nos. 911-914). They comprise the following forms: bFi
(Krepice, hut IX; Rajhrad, hut II), F1 (Krepice, hut II), Ì (Blucina, hut I), X (Blucina,
hut IV; obj. 1; Komorany, hut Z-2; Krepice, hut XIII; Musov, hut II) and =
(Mikulcice, hut IV). Though the swastika and the five parallel lines are more likely to
be symbols, the other three clearly show further examples of BIO f ), C4-6 (£) and
D2 (X). In Krepice, interestingly, three different signs occur on the site and therefore
possibly differentiate between owners of vessels.
A similar picture evolves as to letters on coins. As recent research has shown,
monetary economy was restricted to ‘Celtic’ areas in the 1st half of the 1st c. BC and
was in decline in Rhinehessia after the Gaulish War. “However, by the 20s BC, at the
latest, when coin-use on the Middle Rhine had long gone into recession, the reverse
had happened downstream, and coin-use there seems to have taken a significant
upturn. The bronze rainbow cups . . . which were probably struck in the period after
the Gallic Wars and were still in circulation in large numbers in the Augustan period,
are clearly a Lower Rhine coinage” (D. Wigg 1999: 104). This statement refers to the
Bochum type in particular which displays a variety of single and double letters. N.
Roymans, in an important study recently published (2001), links the Lower Rhine
triquetrum coinage to the ethnogenesis of the Batavi. He argues that the shift of the
centre of production of triquetrum coins from the eastern Middle Rhine region (Lahn
valley, Mardorf type) to the Lower Rhine in the 2nd half of the 1st c. BC (c. 50—15 BC,
Bochum type) is related to the historically documented migration of a group of Chatti
to the (later) territory of the Batavi (Tacitus, Germania 29.1; History 4.12; Roymans
2001: 118). The earliest type a, without extra mark, probably minted in the mid 1st c.
BC (2001: 107), is distributed in the upper Lahn valley (and as far south as the Tëtelbierg in Luxemburg and the middle Main area) as well as the Lower Rhine region
(ibid.: 112). Later emissions with extra marks only sporadically occur east of the
Rhine, except type s, probably connected with movements of the Roman army (ibid.:
131 f). So it is obvious that tribes on the Lower Rhine from c. 50 BC on did not only
use late ‘Celtic’ coins with inscriptions but that they began to mint them by themsel¬
ves. Thus, they must have been acquainted with the alphabet used for coin legends. It
may be remembered that the φ-type occurs several hundred times. The preservation
of the old writing custom is fully in accordance with other aspects of a Batavian cul¬
tural identity of its own, as the maintenance of native traditions of house-building,
the high esteem for warfare and cattle (Roymans 1995) as well as Celtic ideas of poli¬
tical power and legitimacy (Enright 1996: ch. V.l; 8 as to Julius Civilis and Veleda).
All these instances clearly indicate that the La Tène script was already known to
the population in the northern plains around the beginning of our era. They probably
learnt it from residents in the ancient homes of Celtic tribes in southern Germany and
Bohemia or from migrating experts of ‘Celtic coinage’. From these areas, one way
could have been a spread along the river Elbe (cf. Salac 1997; 1999) across the
homeland of the immigrants from the north. It could reach the Liibsow/Lubieszewo
culture in central Germany in particular, which has already yielded evidence of ves¬
sel marks (Sakar 1965: 154; Novotny 1955) and which is also rich in Roman imports.
A second way could have been from the home of inscribed rainbow cups along the
Middle and Lower Rhine to the neighbouring Usipetri, Sugambri, Tencteri etc. and
beyond. Both ways of transmission which are imaginable meet in the lower German
plains of Schleswig-Holstein where, beside Scandinavia, the earliest runic inscrip¬
tions from about AD 150-200 have been found.
These are the 0vre Stabu spearhead (Oppland, Norway), nine objects from Illerup,
including the wooden handle of a lighter and the mouthpiece of a drinking horn (all c. AD
150-200), the Thorsberg chape and shieldboss (Schleswig-Holstein), clasps from Npvling,
Næsbjerg (North and South Jutland), Himlingpje, Værlpse (Sjælland, Denmark), and Gárdlösa (Skáne, Sweden, c. AD 200), the Dahmsdorf (Brandenburg) and Kowel spearheads
(Volhynia, both c. AD 250), a suspension loop, the haft of a hatchet and two arrow shafts
from Nydam (c. AD 250-300), and finds from Vimose (buckle, chape, comb, woodplane,
sheathplate; Fyn, Denmark, AD 200-300). This list is almost complete up to c. AD 300
(Krause 1966: 313 f; Antonsen 1975: 29-76; Stoklund 1995; 1998; Ilkjær 1996; Diiwel
2001: 23-31).
Further evidence is probably provided by ‘proto-runic’ inscriptions from the 1st
half of the 1st c. AD on the disputed Meldorf fibula (Diiwel & Gebühr 1981; Mees
1997) and on a sherd from an excavation at Osterrönfeld in a purely ‘Germanic’ con¬
text (Dietz et al. 1996). Though the interpretation of the letters is difficult, it seems
possible to recognize two forms of <r>, one in the shape f) (Meldorf), the other k
(Osterrönfeld); the Meldorf find also shows vertical strokes, <i>, as well as an almost
illegible letter. B. Mees suggests the reading <ir.il.i.>, dative (?) of e/irilaz ‘(rune)
master’, against Düwel’s <hiwi> ‘for (a female) Hiwi’ and B. Odenstedt’s (1989)
<idin> ‘for Ida/Iddo’. Most significantly, these incisions can be seen as an interme¬
diate stage between the La Tène script and the runes. Firstly, they use variant forms
of <r>, one of them being of the common Norican (and La Tène ?) type; and secondly,
they appear on a Norican fibula type, Almgren II 24—26 (. Rollenkappenfibel , Düwel &
Gebühr 1981: 159), and on a potsherd which is uncommon in view of the later runic
B. Mees (1999) underlined the remarkable similarities the Germanic and Celtic writ¬
ing traditions both have in common. They do not only share the very name of the runes in
the sense of ‘secret’, they also seem to use letters as a means of divination. Mees draws
attention to the noun lidssati-“sorcière opérant avec l’écriture” (Lambert 1994: 166) men¬
tioned several times on the plomb du Larzac. A mantic use of the runes (as well as Ogam
and Semitic Abgad) was also suggested by E. Seebold (1993: 142 f; cf. on runes and
‘magic’, Polomé 1996).
Marcomanni as mediators have already been suggested by C. Marstrander (1928:
120-124; 185). It is well known that Maroboduus “initiated the formation of a large
state” and “is said to have annexed to Boiohaemum all the neighbouring tribes”
(Dobiás 1960: 157). To the north, his empire is said to have included regions in the
Oder valley, northern Brandenburg as far north-west as the lower Elbe, and perhaps
the Pomeranian coast east to the lower Vistula (ibid.). The rest of the Boii who
remained in their ancient home were subjugated by Maroboduus after 9 BC (Tacitus,
Germania 42) and completely amalgamated in the 1st c. AD. Those Boii who migrated
to the middle Danube area during the first half of the first century BC obviously aban¬
doned their native script. In their new home they minted inter alia uniquely large sil¬
ver pieces (called ‘hexadrachms’ by R. Gobi 1994) in Bratislava since the mid sixties
BC (G. Dembski 1999: 55 f contra Gobi 1994: 38 1). The fifteen preserved names of
Boian princes — most familiar are NONNOS and BIATEC (Birkhan 1971; Markey
2001b) — show partly Germanic influence in FARIARIX ‘king of ferries or ferrymen’
and AINORIX ‘sole ruler’ (Birkhan in Gobi 1994: 69 f; 73 f). The inscriptions on these
coins were exclusively in the Latin alphabet, thus paying tribute to the rapidly chang¬
ing political and economical situation in the first century BC (cf. for Magdalensberg,
Markey 2001a: 144).1
Adler, H. & W. Schön 1979. “[Fundchronik] Wien 19— Leopoldsberg”. Fundberichte aus Öster¬
reich 17 (1978): 303-306
Adler, H. & E. Siegmeth 1988. “[Fundchronik] Großrußbach”. Fundberichte aus Österreich 26
(1987): 225-226
Alföldy, G. 1974. Noricum. (The Provinces of the Roman Empire). London & Boston
— 1977. “Die Personennamen in der römischen Provinz Noricum”. Uonomastique latine,
(Paris): 249 ff
— 1993. “Die Personennamen auf den Bleietiketten von Kalsdoif (Steiermark) in Noricum”.
in: F. Heidermanns, H. Rix, E. Seebold (eds.). Sprachen und Schriften des antiken Mittel¬
meerraums. Festschrift für Jürgen Untermann. Innsbruck: 1—32
Allen, D.F. & D. Nash 1980. The coins of the ancient Celts. Edinburgh
Altheim, F. 1950. “Runenforschung und Val Camonica”. Nouvelle Clio 1/2 (no. 4): 166-185
— & E. Trautmann(-Nehring) 1939. Vom Ursprung der Runen. Berlin
— 1942. Kimbern und Runen. Berlin
Antonsen, E.H. 1975. A concise grammar of the older runic inscriptions. Tübingen
— 1982. “Zum Ursprung und Alter des germanischen Futharks”. in: E.S. Dick & K.R. Jankowsky. Festschrift für Karl Schneider. Amsterdam & Philadelphia: 3-15
Arslan, E.A. 2000. “La monetazione con legende leponzie e la monetazione preromana dell’
area leponzia e insubre”, in: Leponti. voi. 2: 223-233
1. The originai version of this paper was read at The Eleventh International Congress of Celtic
Studies, University College Cork, 25—31 July, 1999. Ba mhaith Horn gabháil buiochas mór le heagraithe na Comhdhála gur feidir léamh an pháipéir seo faoi seanscript Cheilteach. My special
thanks go to Dr Norbert Baum, Prof. Natalie Venclová and Prof. Vincent Megaw from whose contri¬
butions in the unique atmosphere of the Congress and afterwards I profited so much and who encou¬
raged me to continue and publish this study; Dr Rosemarie Cordie-Hackenberg, Prof. Patrizia de
Bernardo Stempel, Prof. Joseph F. Eska, Dr Klaus-Peter Goethert, Prof. Pierre-Yves Lambert, Dr
Matthias Leicht, Robert Loscheider MA, Prof. Gemot Piccottini, Kurt Scheuerer, Dr Jürgen Uhlich,
for discussion and information, Paloma Devine for correcting the English, Nadine Quenouille MA
for improving the French summary, Iwona B. Wesolowska MA for her help with texts in Slavonic lan¬
guages, and finally Prof. Venceslas Kruta for accepting the paper in this journal.
Artner, W. 1988/89. “Die provinzialrömischen Gräber von Gleisdorf in der Oststeiermark”.
Mitteilungen der Österreichischen Gesellschaft für Or-und Frühgeschichte 38/39: 83
— 1995. “Ratschendorf: Funde aus dem Gemeindegebiet und Grabungen im norisch-pannonischen Hügelgräberfeld ‘Hügelstaudach’”. (R. Hebert, F.J. Schober etc.). Fundberichte
aus Österreich 33 (1994): 39; 64 pl. 4 no. 2
— 1998/99. “Der Frauenberg bei Leibnitz, Steiermark, in der Spâtlatènezeit und in der vorclaudischen Kaiserzeit”. Archaeologia Austriaca 82/83: 221-341
Asen, K. 1903. “Funde von Steinen mit eingeritzten oder eingegrabenen Zeichen aus Grabhü¬
geln der Hallstattzeit bei Heiligenstadt, Lengau, Ober-Oesterreich”. Prähistorische Blätter
15: 1-8; pl. 1
Bakker, L. & B. Galsterer-Kröll 1975. Graffiti auf römischer Keramik im Rheinischen Lan¬
desmuseum Bonn. (Epigraphische Studien 10). Cologne
Balogh, A. 1927. “Az esztergomi múzeum bélyeges agyagedényeirol”. Archaeologiai Ertesito
(Budapest) 41: 209-217
Bammesberger, A. (ed.) 1991. Old English runes and their continental background. (Anglistische Forschungen 217). Heidelberg; rev.: R. Nedoma. Sprache 35 (1991-1993): 248—250
Barth, F.E. 1984. “Eine prähistorische Stempelzimmerung aus dem Salzbergwerk Hallstatt”.
Archaeologia Austriaca 68: 63—71
— & N. Leutner 1993. “[Fundchronik] KG Hallstatt, MG Hallstatt, YB Gmunden”. Fundbe¬
richte aus Österreich 31 (1992): 461162
Bauer, R. & H. Maurer 1978. in: Fundberichte aus Österreich 16 (1977): 380; 382 fig. 214
Bez. = Bezirk (district, Austria)
Binaghi, M.A. & G. Rocca 1999. “Sesto Calende (VA), loe. Presualdo”. Studi Etruschi 63
(1997): 437-447
Birkhan, H. 1971. “Die ‘keltischen’ Personennamen des boiischen Großsilbers”. Die Sprache
17/1: 23-33
Blanchet, A. 1905. Traité de monnaies gauloises. Paris
Boardman, J. 1974. Athenian blackfigure vases. A handbook. London
Bosman, A.V.A.J. & T. Looijenga 1996. “A runic inscription from Bergakker (Gelderland), The
Netherlands”. Amsterdamer Beiträge zur älteren Germanistik 46: 9-16
Brandt, J. 2001. Jastorf und Latène. Kultureller Austausch und seine Auswirkungen auf soziopolitische Entwicklungen in der vorrömischen Eisenzeit. Rahden/Westfalen
Bren, J. 1956. “Oppidum Holubov - Trisov”. Gasopis Národního Muzea (Prague) 125: 9-16
Bridger, C. & C.v. Camap-Bomheim (eds.) 1997. Römer und Germanen — Nachbarn über
Jahrhunderte (British Archaeological Reports. International Series 678). Oxford
Burkhardt, A. 1998. Quantitative Methoden zur keltischen Numismatik am Beispiel der Münzfunde aus latènezeitlichen Siedlungen der Oberrheinregion. Bern
Cagnat, R. 1914. Cours d’épigraphie latine. (4th ed.). Paris, (repr. Rome 1964)
Camap-Bomheim, C. von. 1998. “Graphit und Graphittonkeramik”. Reallexikon der German¬
ischen Altertumskunde. Berlin & New York. voi. 12: 593-598
Cech, B. 1989. “Mittelalterliche und frühneuzeitliche Keramik aus Tulln, Niederösterreich”.
Archaeologia Austriaca 73: 167-221
Celti = I Celti (catalogue of the exhibition in Palazzo Grassi, Venice). S. Moscati, O.H. Frey, V.
Kruta, B. Raftery, M. Szabó (dir.). Milan 1991
Cizmár, M. 1994. “Ein Beitrag zur Kenntnis der Herstellung der spâtlatènezeitlichen Graphit¬
keramik in Südmähren”. Acta Musei Moraviae. Casopis Moravského Muzea 79: 85—93
— 1995. “Znacky na dnech laténské keramiky na Morave II” Acta Musei Moraviae. Casopis
Moravského Muzea 80: 109—120
— 1996. “Neue Erkenntnisse über die Spâtlatènezeit in Mähren”, in: Jerem 1996: 359—369
Cizmár, M. & J. Meduna 1985. “Bodenzeichen auf latènezeitlicher Keramik in Mähren —
Znacky na dnech laténské keramiky na Morave”. Památky Archeologické 76: 78—100
Ciurletti, G. & F. Marzatico (eds.) 1998. I Reti / Die Räter. Atti del simposio 23-25 settembre
1993 Castello di Stenico, Trento. (Archeoalp. Archeologia delle Alpi 5). Trento (pubi. 1999)
Colbert de Beaulieu, J.-B. 1956. “Notes sur le monnayage épigraphe du Norique”. Ogam 8:
— 1965. “Légendes monétaires de Gaule en caractères dérivés de’l Etrusque”, in: Actes du
90e Congrès national des Sociétés Savants. Nice: 445-450
— 1965a. “Remarques sur les inscriptions monétaires nord-italiques en Gaule cisalpine et en
Gaule transalpine”, in: Congresso internazionale di numismatica, Roma 1961. voi. 2.
Rome: 179-184
Cornell, T. 1991. “The tyranny of the evidence: a discussion of the possible uses of literacy in
Etruria and Latium in the archaic age”, in: M. Beard, A.K. Bowman etc. (eds.). Literacy in
the Roman world. (Journal of Roman archaeology, suppl. ser. 3). Ann Arbor: 7-33
Corthals, J. 1999. “Zur gallischen Inschrift von St.-Germain-Sources-Seine”. Historische
Sprachwissenschaft 112: 101
Creighton, J.D. & R.J.A. Wilson (eds.) 1999. Roman Germany. Studies in cultural interaction,
(Journal of Roman archaeology, suppl. ser. 32). Portsmouth, RI
Cuscito, G. (ed.) 2001. 1 Celti nell’alto adriatico. (Antichità altoadriatiche 48). Trieste
DAG = J. Whatmough, The dialects of ancient Gaul. Prolegomena and records of the dialects,
Cambridge MA 1970
De Bernardo Stempel, P., 2002. « Die sprachliche Analyse Keltischer Theonyme », Zeitschrift
für celtische Philologie 53:41-69.
De la Tour = H. de la Tour / B. Fischer, Atlas de monnaies gauloises, (2nd ed.), Paris 1994
Dembski, G., 1999. “Keltische Münzen in Ostösterreich— Boier und Burebista”, Acta Musei
Napocensis (Cluj) 36/1, 51-58
Derolez, R.L.M. 1998a. The origin of the runes: an alternative approach. (Academiae Analecta,
Klasse der Letteren 60/1). Brussels
— 1998b. “On the ‘otherness’ of the Anglo-Saxon runes and the ‘perfect fit’ of the Futhark” in:
Düwel 1998: 102-116
Dietz, K. 2000. “Noricum”, in: Der Neue Pauly Vili. H. Cancik & H. Schneider (eds.), Stut¬
tgart & Weimar, col. 1003-1007
Dietz, M., E. Marold & H. Jöns 1996. “Eine frühkaiserzeitliche Scherbe mit Schriftzeichen aus
Osterrönfeld, Kr. Rendsburg-Eckemförde”. Archäologisches Korrespondenzblatt 26: 179—188
DIL = Dictionary of the Irish language. Based mainly on Old and Middle Irish materials, (com¬
pact ed.). Dublin 1983 (repr. 1990)
DLG = X. Delamarre. Dictionnaire de la langue gauloise. Paris 2001 (2nd ed. 2003)
Dobesch, G. 1980. Die Kelten in Österreich nach den ältesten Berichten der Antike. Das norische
Königreich und seine Beziehungen zu Rom im 2. Jh.v.Chr. Vienna etc.
— 1992. “Die Kelten als Nachbarn der Etrusker in Norditalien”, in: L. Aigner-Foresti (ed.).
Etrusker nördlich von Etrurien. Vienna: 161-178
— 1997. “Zu Virunum als Namen der Stadt auf dem Magdalensberg und zu einer Sage der
kontinentalen Kelten” Corinthia 1. 187: 107-128
Dobiás, J. 1960. “King Maroboduus as a politician”. Klio 38: 155-166
Drda, P. & A. Rybová 1995. Les Celtes de Bohême. Paris
Droberjar, E. 1997. Studien zu den germanischen Siedlungen der älteren römischen Kaiserzeit in
Mähren. (Fontes Archaeologici Pragenses 21). Prague
Duval, P.-M. 1989. “Les Gaulois savaient écrire”, in: id. Travaux sur la Gaule (1946—1986).
vol. 1. Rome: 191—197
Düwel, K. 1998. (& coed. S. Nowak). Runeninschriften als Quellen interdisziplinärer Forsch¬
ung. Proceedings of the 4th International Symposium on Runes and Runic Inscriptions in
Göttingen, 4-9 August 1995. (ErgRGA 15). Berlin & New York
— 2001. Runenkunde. (3rd ed.). Stuttgart & Weimar
— & M. Gebühr 1981. “Die Fibel von Meldorf und die Anfänge der Runenschrift”. Zeits¬
chrift für deutsches Altertum und deutsche Literatur 110: 159—175 (with a contr. by
H. Hammon)
Eck, W. 1998. “Inschriften auf Holz. Ein unterschätztes Phänomen der epigraphischen Kultur
Roms”, in: P. Kneissl & V. Losemann (eds.). Imperium Romanum. Studien zu Geschichte
und Rezeption. Festschüft für Karl Christ zum 75. Geburtstag. Stuttgart: 203-217
Egg, M. 1996. Das hallstattzeitliche Fürstengrab von Strettweg bei Judenburg in der Ober¬
steiermark. (Römisch-Germanisches Zentralmuseum, Monograph 37). Mainz
Egger, R. 1953. “Die Ausgrabungen auf dem Magdalensberg 1951. Die Inschriften”. Carinthia
1 143: 70-90
— 1959. “Die Ausgrabungen auf dem Magdalensberg 1956 und 1957. Die Inschriften”.
Carinthia 1 149: 135 ff
— 1968. “Inschriften auf Eß-und Trinkgeschirr vom Magdalensberg”, in: Pro Augusta Rau¬
rica (ed.). E. Schmid, L. Berger, P. Bürgin (reds.). Provincialia. Fs Rudolf Laur-Belart.
Basel & Stuttgart: 269-277
— 1968a. “Zum vorlateinischen Alphabet der Noriker”. Arheoloski Vestnik 19: 37— 42
— (f) 1969. “Die Ausgrabungen auf dem Magdalensberg 1965 bis 1968. Die Inschriften”.
Carinthia I. 159: 365-409 (part of excavation report 12 by H. Vetters & G. Piccottini. ibid.
Eggers, HJ. 1955. “Zur absoluten Chronologie der römischen Kaiserzeit im Freien German¬
ien”. Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 2: 196-244
Ehrenhöfer, J. & E. Pichler 2001. “Spâtlatènezeitliche und römische Funde aus Wien 3, Rud¬
olfstiftung”. Fundort Wien. Berichte zur Archäologie 4: 280-293
Ehrenreich, S., 1994. “Das Gräberfeld von Katsch in der Steiermark. Neuaufnahme der
Funde”, Fundberichte aus Österreich 32 (1993), 9—35
Eichner, H., 1989. “Damals und heute: Probleme der Erschließung des Altkeltischen zu Zeußens Zeit und in der Gegenwart”, in: B. Forssman (ed.), Erlanger Gedenkfeier für Johann
Kaspar Zeuß, (Erlanger Forschungen A49), Erlangen
— , 1994. “Lesung und Deutung der Inschrift”, in: H. Eichner, J. Istenic, M. Lovenjak, »Ein
römerzeitliches Keramikgefäß aus Ptuj (Pettau, Poetovio) in Slowenien mit Inschrift in
unbekanntem Alphabet und epichorischer (vermutlich keltischer) Sprache«, Arheoloski
Vestnik 45, 132-138 (complete article 131-142)
Elboume, R, 2000. “Plain voiceless stop plus laryngeal in Indo-European”. Historische
Sprachwissenschaft 113: 2-28
Enright, M.J. 1996. Lady with a mead cup. Ritual, prophecy and lordship in the European warband from La Tène to the Viking Age. Dublin
Eska, J.F. 1998. “Tau Gallicum”. Studia Celtica 32: 115—127
— & R.E. Wallace 1999. “The linguistic milieu of *Oderzo 7”. Historische Sprachwissenschaft
112: 122-136
— & M. Weiss 1996. “Segmenting Gaul. TOMEDECLAI”. Studia Celtica 30: 289—292
Etrusker 1992. Die Etrusker und Europa. (Altes Museum Berlin, 28.2.— 31.5.1993). M. Pallottino etc. (eds.). Gütersloh & Munich
Etzel, C. 1990. Die Grabungen 1987 und 1988 im keltischen Oppidum Alkimoennis bei Kel¬
heim. (unpubl. MA thesis). Erlangen-Nürnberg
EWAi = M. Mayrhofer. Etymologisches Wörterbuch des Altindoarischen I. Heidelberg 1990
Fansa, M. & R. Schneider 1993. “Die Bohlenwege bei Ockenhausen/Oltmannsfehn, Gde.
Uplengen, Ldkr. Leer”. Archäologische Mitteilungen aus Nordwestdeutschland 16: 23-43
(see also ibid. 18 [1995]: 22)
Fellmann, R. 1991. “Die Zinktafel von Bem-Thormebodewald und ihre Inschrift”. Archäologie
der Schweiz 14/2: 270-273
— 1999 (with W.B. Stern, A. Burkhardt, T. Rehren). “Das Zinktäfelchen vom Thormebodewald
auf der Engehalbinsel bei Bern und seine keltische Inschrift”. Archäologie im Kanton Bern
4B: 133-175
Fischer, F. 1999. Review of S. Rieckhoff 1995. Germania 77/1: 382-386
Fogolari, G. & A.L. Prosdocimi 1988. 1 Veneti antichi. Lingua e cultura, (with contributions by
M. Gamba, A. Marinetti). Padova
Forrer I = R. Forren Keltische Numismatik der Rhein-und Donaulande. Straßburg 1908 (repr.
Graz 1968)
Forrer II = K. Castelin (red.). Forrer. voi. 2: Bibliographische Nachträge und Ergänzungen,
Graz 1969
— 1910. “Die keltogermanischen Triquetrumgepräge der Marser, Sugambrer, Tenkterer und
Ubier”. Jahrbuch der Gesellschaft für lothringische Geschichte und Altertumskunde 22:
Franz, L. 1942. Eine keltische Niederlassung in Südböhmen. (Abhandlungen der Deutschen
Akademie der Wissenschaften Prag 2). Prague
Freeman, P. 2001. The Galatian language. Lampeter etc.
Furtwängler, A.E. 1978. Monnaies grecques en Gaule. Le trésor d’Auriol et le monnayage de
Massalia 525/520-460 αν. J. -G.. Fribourg
Gambacurta, G. & A. Ruta Serafini 2001. “I Celti e il Veneto: appunti per una revisione”, in:
Cuscito 2001: 187-201
Gambari, F.M. & G. Colonna 1988. “Il bicchiere con iscrizione arcaica da Castelletto
Ticino e l’adozione della scrittura nell’ Italia nord-occidentale”. Studi Etruschi 54:
— & G. Kaenel 2001. “L’iscrizione celtica sulla perla da Miinsingen: una nuova lettura”.
Archäologie der Schweiz / Archéologie suisse 24/4: 34—37
Goldenberg, G. 2002. “Bronzezeitlicher Kupferbergbau in Nordtirol”. Archäologie in Deutsch¬
land 3/2002: 58-63
Gleirscher, P. 1996. “Die Kelten im Raum Kärnten aus archäologischer Sicht — Ein For¬
schungsstand”. in: Jerem etc. 1996: 255—266
Gobi, R. 1973. Ostkeltischer Typenatlas. Braunschweig
— 1989. Der norische Tetradrachmenfund 1972 aus Haimburg in Kärnten: Versuch einer
Gesamtrekonstruktion. (Sitzungsberichte der Österreichischen Akademie der Wissenschaf¬
ten. Hist.-phil. Klasse 522). Vienna
— 1994. Die Hexadrachmenprägung der Groß-Boier. Ablauf, Chronologie und historische Rele¬
vanz für Noricum und Nachbargebiete, (with historical, linguistic and epigraphic excursus
by G. Dobesch, H. Birkhan and M. Hainzmann). Vienna
Gosden, C.H. 1984. “Bohemian Iron Age chronologies and the sériation of Radovesice”. Ger¬
mania 62: 289-309
GPC = Geiriadur Prifysgol Cymru. A dictionary of the Welsh language, vol. 1. Cardiff 1950
GPN = Ellis Evans, D. Gaulish personal names. A study of some Continental Celtic formations.
Oxford 1967
Graue, J. 1974. Die Gräberfelder von Ornavasso. Eine Studie zur Chronologie der späten Latèneund frühen Kaiserzeit. Hamburg
Green, M. 1989. Symbol and image in Celtic religious art. London & New York
Griffiths, A. 1999. “The fufiark (and ogam): order as a key to origin”. Indogermanische For¬
schungen 104: 164—210
Grünert, H. 1992. “Keltisch-germanische Kontakte im sächsisch-thüringischen Mittelgebirgsraum und ihre Bedeutung für die ökonomisch-soziale Entwicklung der Germanen”, in:
Kelt.-germ. Besiedlung: 140—152
Hachmann, R. 1993. “Verzierte Lanzenspitzen der Jüngeren Kaiserzeit. Die Vor-und Frühge¬
schichte und die Runeninschriften im älteren FuJ)ark”. in: Kulturen zwischen Ost und West.
Das Ost-West-Verhältnis in vor-und frühgeschichtlicher Zeit und sein Einfluß auf Werden
und Wandel des Kulturraums Mitteleuropa (Fs G. Kossack). A. Lang, H. Parzinger, H. Küs¬
ter (eds.). Berlin: 327-423
— 1994/95. “Goten und Runen. Die gotischen Runeninschriften und die Schrift des Ulfilas”.
Dacia (Bucarest) NS. 38/39: 163-183
Hahnei, B. 1994. “Funde der mittleren Bronzezeit, der älteren Umenfelderzeit sowie der Spätlatène-und Römerzeit in Wien 23, Sulzengasse. Archäologischer Fundbericht”. Archaeologia Austriaca 78: 29-56
Hainzmann, M. 1987. “Die sogenannten Neubürger der ersten Generation in Noricum. Zum
Namenstypus Ti. Iulius Adgelei f. Buccio”. Tyche 2: 29-39
— 1996. “Die Kelten vom Magdalensberg— eine onomastische Bestandsaufnahme”, in: Jerem
etc. 1996: 447-459
Hebert, B. 1991. “Neufunde aus der Steiermark”. Specimina nova dissertationum ex Instituto
Historico Universitatis Quinqueecclesiensis de lano Pannonio Nominatae. Pees: 281—292
Hell, Μ. 1922. “Bodenzeichen auf Gefäßen der Spâtlatènezeit”. Wiener Prähistorische Zeit¬
schrift 9: 109 ff
— 1934. “Bodenzeichen auf keltischen Gefäßen aus Hallstatt”. Germania 18: 189-193
— 1935. “Keltische Töpferzeichen”. Forschungen und Fortschritte 11/14: 182-183
— 1936. “Alte und neue Funde aus Hallstatt”. Mitteilungen der Anthropologischen Gesell¬
schaft in Wien 66: 47-68
— 1940. “Zur Verbreitung der Bodenzeichen auf spätkeltischen Kammstrichgefäßen”. Ger¬
mania 24: 118-123
— 1949. “Keltisches Bodenzeichen auf römischem Gefäßen [.sie] aus Hallstatt, O.-O.”.
Archaeologia Austriaca 2: 81—83
— 1952. “Die Kleinfunde von der Dammwiese in Hallstatt aus den Jahren 1936-1937”. Arch¬
aeologia Austriaca 11: 71-88
Herrmann, W.A. 2000. “Archäologie und Naturwissenschaften - kein Widerspruch”, in:
Archäologie im Landkreis Freising : 13—22
Hoppe, M. 1982. “Neue Siedlungsfunde der Bronze-und Eisenzeit aus dem Taubergrund”.
Fundberichte aus Baden-Württemberg 7: 73-207
Hörler, H. 1953. Die ur- und frühgeschichtliche Besiedlung des Bezirkes Gänserndorf, Nied¬
erösterreich, (unpubl. PhD thesis), Vienna (non vidi )
Horakova-Jansova, L. 1955. “Latenska tuhova keramika v Cechach a na Morave”, Památky
Archeologické 46, 1 34—184
IEW = J. Pokomy. Indogermanisches etymologisches Wörterbuch. 2 vol. Bern 1959-1969
Ilkjær, J. 1996. “Runeindskrifter fra mosefund i Danmark - kontekst og oprindelse”. in: Looijenga & Quak 1996: 63—75
Istenic, J. 1985. “Zapisi v venetski pisavi na Koroskem in v Sloveniji”, Zgodovinski Casopis
39: 313-334
Jablonka, P. 1993. “Die venetischen Inschriften und die eisenzeitliche Besiedlung der Gurina
bei Dellach im Gailtal, Kärnten”, Archäologie Österreichs. Mitteilungen der Österreichis¬
chen Gesellschaft für Ur-und Frühgeschichte 43. 4/1: 4-9
— 2001. Die Gurina bei Dellach im Gailtal. (Aus Forschung und Kunst 31). Klagenfurt
Jacobi, G. 1974. “Zum Schriftgebrauch in keltischen Oppida nördlich der Alpen”. Hamburger
Beiträge zur Archäologie, voi. 4. Wolfgang Dehn zum fünfundsechzigsten Geburtstag am 6.
Juli 1974 gewidmet. Hamburg: 171—181
Jahn, M. 1919. “Der Spätlatene-Fund von Tschiläsen Kr. Guhrau”. Mannus 10 (1918): 15-24;
pi. 1
— , 1937. “Rätische Kultureinflüsse in Böhmen und Schlesien während des letzten vorchrist¬
lichen Jahrhunderts”, Sudeta 13, 103—111
Jerem, E. & J. Kardos 1985. “Entwicklung und Charakter der eisenzeitlichen Graphitton¬
ware”. Mitteilungen der österreichischen Arbeitsgemeinschaft für Ur- und Frühgeschichte
35: 65-75
Jerem, E., (et al.) 1996. Die Kelten in den Alpen und an der Donau, Akten des Internationalen
Symposions St. Pölten, 14.-18. Oktober 1992. E. Jerem, A. Krenn-Leeb, J.-W. Neugebauer,
O.H. Urban (eds.). Budapest & Vienna
Jordán Cólera, C. 1998. Introducción al Celtibérico, Zaragoza (rev. D. Stifter. Die Sprache 40/1
[1998]: 128-132)
Kappel, I. 1969. “Die Bodenzeichen der kammstrichverzierten Graphittontöpfe”, in: ead. Die
Graphittonkeramik von Manching. (Die Ausgrabungen in Manching 2). Wiesbaden: 104122; 188-191
— 1976. “Der Münzfund von Mardorf und andere keltische Münzen aus Nordhessen”. Ger¬
mania 54: 75-101
Kelly, F. 1988. A guide to early Irish law. (Early Irish law series 3). Dublin (repr. 1991)
Kelt.-germ. Besiedlung = Beiträge zur keltisch-germanischen Besiedlung im Mittelgebirgsraum.
Internationales Kolloquium 15.— 17. Mai 1990 in Weimar, (ed.) Thüringisches Landesamt
für archäologische Denkmalpflege. (Weimarer Monographien zur Ur-und Frühgeschichte
28). Stuttgart 1992
Keltjt. = H. Dannheimer & R. Gebhard (eds.). Das keltische Jahrtausend. Munich & Mainz
Kelten in Mitteleuropa = Die Kelten in Mitteleuropa. Ausstellungskatalog Hallein 1980.
Salzburg 1980
KGPN = K.H. Schmidt. Die Komposition in gallischen Personennamen, (orig, in Zeitschrift für
celtische Philologie 26). Tübingen 1957
Klindt-Jensen, 0. 1949. “Foreign influences in Denmark’s early Iron Age” . Acta Archaeologica
(Copenhagen) 20: 1-229 (esp. p. 97 fig. 66)
Kostial, M. 1997. Kelten im Osten. Gold und Silber der Kelten in Mittel-und Osteuropa. Samm¬
lung Lanz. Munich
Krämer, W. 1982. “Graffiti auf Spâtlatènekeramik aus Manching” Germania 60: 489-499
— schichtsblätter
1984. “Ein keltisches
49: 293-294
Schriftzeugnis vom Dürmberg bei Hallein?”. Bayerische Vorge¬
— 1996. “Figürliche Ritzzeichnungen auf Gefäßscherben Glatter Drehscheibenkeramik der
Mittel-bis Spâtlatènezeit”. Germania 74: 361-376
Krause, W. 1937. Runeninschriften im älteren Futhark. Halle (Saale)
— 1943. “Zum Stand der Runenforschung”. Göttinger Gelehrte Anzeigen 205 (no. 7—8): 231—
— 1966 (with H. Jankuhn). Runeninschriften im älteren Futhark. 2 vols. Göttingen
Kruta, V. 1991. “La scrittura / Celtic writing” in: Celti : 491-497
— 2000. Les Celtes. Histoire et dictionnaire. Des origines à la romanisation et au christianisme.
Kt. = Kanton (district, Switzerland)
Lambert, R-Y. 1992. “Diffusion de l’écriture gallo-grecque en milieu indigène”, in: Marseille
grecque et la Gaule. Actes du Colloque international d’ Histoire et d' Archéologie et du Ve
Congrès archéologique de Gaule méridionale (Marseille, 18-23 novembre 1990). M. Bats,
G. Bertucchi etc. (eds.). Lattes & Aix-en-Provence: 289—294
— 1994. La langue gauloise, description linguistique, commentaire d’inscriptions choisies. Paris
(later eds.: 1995; 1997; in print)
— et al. 2002. “Dossier: Parlez-vous gaulois?”. L’archéologue/archéologie nouvelle 59: 4-27
Lang, A. & U. Schultz 1995. “Ein hallstattzeitlicher Brandopferplatz bei Farchant”. Dos arch¬
äologische Jahr in Bayern: 74-77
Lasch, H. 1999. Berching-Pollanten I. Die Keramik der spätkeltischen Siedlung Berching-Pollanten, Lkr. Neumarkt i.d. Oberpfalz. (Internationale Archäologie, 60). Rahden/Westf.
LEIA = J. Vendryes. Lexique étymologique de l’irlandais ancien. 7 vols. Dublin & Paris 1959Leicht, M. 2000. Die Wallanlagen des Oppidums Alkimoennis/Kelheim. Zur Baugeschichte und
Typisierung spätkeltischer Befestigungen. Rahden/Westf.
Leicht, M. & B. Ziegaus 2000. “Ein Münzschatzfund aus dem Oppidum Manching”. Das
archäologische Jahr in Bayern 1 999 : 40-43
Lejeune, M. 1974. Manuel de la langue Vénète. (Indogermanische Bibliothek). Heidelberg
— 1989. “Un abécédaire corinthien du Ve siècle en Dardanie”. Kadmos 28: 14—18
— 1991. “Venetica XIX. L’écriture vénète à la lumière des documents pannoniens”. Latomus
50: 785-797
Lenerz-de Wilde, M. 1987. “Zur Funktion der Schrift in der Frühgeschichte Mittel- und
Nordeuropas”. Acta Praehistorica et Archaeologica 19: 85—92
Lengyel, I. 1968. “Früheisenzeitliche inschriftliche Denkmäler”. Acta Antiqua Academiae
Scientiarum Hungaricae 16: 105—116
Leponti = R.C. de Marinis & S. Biaggio Simona (eds .). I Leponti tra mito e realtà. 2 voi. Locamo
Liechtensteinisches Landesmuseum Vaduz (ed.) 1975. Wegleitung, Vaduz
Lkr. = Landkreis (district, Germany)
Lochner v. Hüttenbach, F. 1989. Die römerzeitlichen Personennamen der Steiermark : Herkunft
und Auswertung. (Arbeiten aus der Abt. Vergleich. Sprachwissenschaft, 2). Graz
Looijenga, T. & A. Quak 1996. Frisian runes and neighbouring traditions. Proceedings of the
first international symposium on Frisian runes at the Fries Museum, Leeuwarden 1994.
{Amsterdamer Beiträge zur älteren Germanistik, 45). Amsterdam & Atlanta GA
Lund, A.A. 1998. Die ersten Germanen. Ethnizität und Ethnogenese. Heidelberg
Maier, F., {et al.) 1992. Ergebnisse der Ausgrabungen 1984-1987 in Manching. F. Maier, U.
Geilenbrügge, E. Hahn, H.-J. Köhler, S. Sievers (eds.). (Ausgrabungen in Manching 15).
— 1996/97. “Das nordmainische Hessen im Randbereich der keltischen Oppida-Kultur”.
Berichte der Kommission für archäologische Landesforschung in Hessen 4: 9-17
Mayer, E.F. 1976. “Zur Herkunft der Marken auf umenfelder-und hallstattzeitlichen Bronze¬
geräten des Ostalpenraumes”. Germania 54: 365-381
Mancini, A. 1998. “Iscrizioni retiche: aspetti epigrafici”, in: Ciurletti & Marzatico 1998: 297333
Marichal, R. 1988. Les graffites de La Graufesenque. Paris
Marinetti, A. 1992. “Documenti epigrafici da S. Lorenzo di Sebato (Val Pusteria, Bolzano)”, in:
I.R. Metzger & P. Gleirscher (eds.). Die Räter - 1 Reti, Bolzano: 691—700
— 1998. “Il venefico: bilancio e prospettive”, in: ead., M.T. Vigolo & A. Zamboni (eds.).
Varietà e continuità nella storia linguistica del Veneto. Atti del Convegno della Società Ital¬
iana di Glottologia, Padova - Venezia, 3-5 ottobre 1996. Rome: 49—99
Marinetti, A. & A.L. Prosdocimi 1994. “Le legende monetali in alfabeto leponzio”. in: Numis¬
matica e archeologia del celtismo padano. Atti del convegno internazionale (Saint-Vincent)
Aosta: 2318
Markey, T. 2000. “Rheto-Celtic PN AKE-VITAMULATE = *BEN(N)ACOS WINDAMOLATOS”. General Linguistics 37/1 (1997): 37-10
— 2001a. “A tale of two helmets: the Negau A and B inscriptions”. Journal of Indo-European
Studies 29: 69-172
— 2001b. “Boiian numismatic BIATEC”. Historische Sprachwissenschaft 114: 307—308
Marstrander, C J.S. 1928. “Om runene og runenavnenes oprindelse”. Norsk Tidskrift for
Sprogvidenskap 1: 85—188
Maurer, H. 1980. “[Fundchronik] Roseldorf”. Fundberichte aus Österreich 18 (1979): 411—13
— &440-441
F. Strouhal 1982. “[Fundchronik] Ladendorf”. Fundberichte aus Österreich 20 (1981):
McCone, K. 1996. Towards a relative chronology of ancient and medieval Celtic sound change
(Maynooth Studies in Celtic linguistics 1). Maynooth
Meduna, J. 1980. Die latènezeitlichen Siedlungen in Mähren. Prague
Mees, B. 1997. “A new interpretation of the Meldorf fibula inscription”. Zeitschrift für deut¬
sches Altertum und deutsche Literatur 126: 131-139
— 1999. “The Celts and the origin of the runic script”. Studia Neophilologica 71: 143—155
Meid, W. 1989. Zur Lesung und Deutung gallischer Inschriften. (Innsbrucker Beiträge zur
Sprachwissenschaft. Vorträge und kleinere Schriften 40). Innsbruck
— 1999. “Forschungsbericht Altkeltische Sprachen II”. Kratylos 44: 1-19
— 2000. “Forschungsbericht Altkeltische Sprächen III”. Kratylos 45: 1-28
Meyer, M. 1994/95. “Die Siedlung der Spâtlatène-und Kaiserzeit von Mardorf, Fdst. 23. Vor¬
bericht über die Ausgrabungen 1993-1994”. Berichte der Kommission für Archäologische
Landesforschung in Hessen 3: 47-58
MLH = J. Untermann (ed.). Monumenta Linguarum Hispanicarum, vol. 1—4. Wiesbaden 1975—
Mohen, J.-R, A. Duval, C. Eluère (eds.) 1987. Trésors des Princes Celtes. Paris
Morandi, A. 1998. “Epigrafia camuña. Osservazioni su alcuni aspetti della documentazione”.
Revue belge de philologie et d’histoire 76: 99-124
Morris, R.L. 1988. Runic and Mediterranean epigraphy. Odense
Morton, F. 1958. “Spâtlatènekeramik aus Hallstatt, Oberösterreich”. Archaeologia Austriaca
24: 35-38
Moßler, G.M. 1961. “Sinnzeichen auf einheimischer Keramik”, in: Bericht über den V. Interna¬
tionalen Kongreß für Vor- und Frühgeschichte, Hamburg vom 24. bis 30. August 1958. G.
Bersu (ed. with W. Dehn). Berlin: 557—561
— 1986. “Die bodenständige Gefäßkeramik der Ausgrabung 1978”. in: H. Vetters & G. Piccottini. Die Ausgrabungen auf dem Magdalensberg 1975-1979. (Magdalensberg Grab¬
ungsbericht 15). Klagenfurt: 173-174
Motta, P. 2000. “La documentazione epigrafica e linguistica”, in: Leponti. voi. 2: 181-222
Nedoma, R. 1995. Die Inschrift auf dem Helm B von Negau. Möglichkeiten und Grenzen der
Deutung norditalischer epigraphischer Denkmäler. (Philologica Germanica 17). Vienna
— 1998. “Zur Problematik der Deutung älterer Runeninschriften - kultisch, magisch oder
profan?”, in: Düwel 1998: 24-54
— & H. Reichert 1998. “Ergebnis einer Untersuchung der Inschrift auf Helm B von Negau
mittels Auflichtmikroskop”. Die Sprache 40/2: 224-225
Nemeskalova-Jiroudkova, Z. 1998. Keltsky poklad ze Starého Kolina. Vysehrad
Nickel, C. 1999. Gaben an die Götter. Der gallo-römische Tempelbezirk von Karden (Kr.
Cochem-Zell, D). Montagnac
Novotny, B. 1955. “Hrob velmoze z pocátku doby rímské v Praze-Bubenci”. Pamdtky
Archeologické 46: 227—257
Odenstedt, B. 1989. “Further reflections on the Meldorf inscription”. Zeitschrift für deutsches
Altertum und deutsche Literatur 118: 77—85
— 1990. On the origin and early history of the runic script. Typology and graphic variation in
the older futhark. (Acta Academiae Regiae Gustavi Adolphi 59). Uppsala
okr. = okres (district, Czech and Slovak Republics)
OxCal 2000. Oxford University Research Laboratory for Archaeology and the History of Art:
Calibration program OxCal 3.5 of 19.07.2000 at <http://www.rlaha.ox.ac.uk/orau/06_01.
Pauli, L. 1992. “«Fremdstücke» im rätisehen Raum. Überlegungen zu ihrem handeis-, wirtschafts-und kulturgeschichtlichen Aussagewert”, in: Räter 1992: 611-629
Pescheck, C. 1939. Die frühwandalische Kultur in Mittelschlesien. Leipzig
— 1976. “Bodenzeichen auf keltischen Töpfen in Nordbayem”. Mitteilungen der Gesellschaft
für Salzburger Landeskunde 115 (1975): 369—371
— 1977. “Ausklang der Latène-Zivilisation und Anfänge der germanischen Besiedlung in
Nordbayem”. in: B. Chropovsky (ed.). Ausklang der Latène-Zivilisation und Anfänge der
germanischen Besiedlung im mittleren Donaugebiet. Symposium Malé Vozokany 1972:
— 1978. Die germanischen Bodenfunde der römischen Kaiserzeit in Mainfranken. (Münchner
Beiträge zur Vor-und Frühgeschichte 27). Munich
Piccottini, G. 1989. Bauen und Wohnen in der Stadt auf dem Magdalensberg. (Österreichische
Akademie der Wissenschaften, Phil.-hist. Klasse, Denkschrift 208). Vienna
PID = J. Whatmough. The Prae-Italic Dialects of Italy, vol. 2. Cambridge MA 1933
Pingel, V. 1971. Die Glatte Drehscheibenkeramik von Manching. (Ausgrabungen Manching 4).
Pink 1974. Die Münzprägung der Ostkelten und ihrer Nachbarn. (2nd ed. with R. Gobi). Braun¬
schweig. (1st ed. Budapest 1939)
Pittioni, R. 1942. “Zur Frage nach der Herkunft der Runen und ihrer Verankerung in der Kul¬
tur der europäischen Bronzezeit”. Beiträge zur Geschichte der deutschen Sprache und Lite¬
ratur 65: 373-384
— 1950. “Zur zeitlichen Stellung der Kerbhölzer von der Kelchalpe bei Kitzbühel, Tirol”.
Nouvelle Clio 1/2 (no. 7): 385-387
Polomé, E. 1996. “Beer, runes and magic”. Journal of Indo-European Studies 24: 99-105
Preuschen, E. & R. Pittioni 1954. “Untersuchungen im Bergbaugebiet Kelchalm bei Kitz¬
bühel, Tirol (3. Bericht)”. Archaeologia Austriaca 15: 1-97
Prosdocimi, A.L. 1989. “La trasmissione dell’alfabeto in Etruria e nell’Italia antica: insegna¬
mento e oralità tra maestri e allievi”, in: Atti del secondo congresso internazionale etrusco,
Firenze 1985. (red. G. Maetzke). Rome. voi. Ill: 1321—1369
— 1991. “Note sul celtico in Italia”. Studi Etruschi 57: 139-177
— 1995. Filoni indoeuropei in Italia. Riflessioni e appunti = A. Landi (ed.). U Italia e il Med¬
iterraneo antico: atti del convegno della Società Italiana di Glottologia. Fisciano - Amalfi Raito 1993. voi. 3. Pisa
PSS = Prähistorische Staatssammlung, Munich
Reitzenstein, W.-A. Frhr. v. 1995. “Die Gräzisiemng bayerischer Ortsnamen”, in: M. Ofitsch &
C. Zinko (eds.). Studia Onomastica et Indogermanica. Fs Lochner von Hüttenbach. Graz:
Radke, G. 1967. “Die italischen Alphabete”. Studium, Generale 20/7: 401-431
Räter = LR. Metzger & P. Gleirscher (eds.). Die Räter / 1 Reti , (Schriftenreihe der Arbeits¬
gemeinschaft Alpenländer). Bolzano 1992
Reichert, H. 1987. Thesaurus Palaeogermanicus 1/1: Lexikon der altgermanischen Namen,
Text. Vienna
RGA XII. Reallexikon der Germanischen Altertumskunde von J. Hoops. (2nd ed.), vol. XII:
Getränke - GreiftierStil. Berlin & New York 1998
RGA XVI. Reallexikon der Germanischen Altertumskunde von J. Hoops. (2nd ed.), vol. XVI:
Jadwingen - Kleindichtung. Berlin & New York 2000
Rieckhoff, S. 1995. Süddeutschland im Spannungsfeld von Kelten, Germanen und Römern.
(Trierer Zeitschrift. Beiheft 19). Trier
RIG II/l. Recueil des inscriptions gauloises, vol. II fase. 1: Les textes gallo-étrusques. Les textes
gallo-latins sur pierre. (Gallia suppi. 45). Paris 1988
RIG IV. Recueil des inscriptions gauloises, vol. IV: Les légendes monétaires. J.-B. Colbert de
Beaulieu (f) & B. Fischer. (Gallia suppi. 45). Paris 1998
Rix, H. 1992. “Thesen zum Ursprung der Runenschrift”, in: L. Aigner-Foresti (ed.). Etrusker
nördlich von Etrurien. (SÖAW 589). Vienna: 41 1 —44]
— 1997. “Germanische Runen und venetische Phonetik”, in: T. Birkmann, H. Klingenberg, D.
Nübling, E. Ronneberger-Sibold (eds.). Vergleichende germanische Philologie und Skandinavistik. Festschrift für Otmar Werner. Tübingen: 231-248
— 1997a. “II problema del retico”, in: A. Marinetti, M.T. Vigolo, A. Zamboni. Varietà e con¬
tinuità nella storia linguistica del Veneto. (Atti del Convegno della Società Italiana di Glot¬
tologia 1996). Rome: 25-48
— 1998: see Ziegaus, B. & H. Rix 1998
Römer-Martijnse, E. 1990. Römerzeitliche Bleietiketten aus Kalsdof, Steiermark. (Öster¬
reichische Akademie der Wissenschaften. Denkschriften 205). Vienna
Roymans, N. 1995. “Romanization, cultural identity and the ethnic discussion. The integration
of Lower Rhine populations in the Roman Empire”, in: J. Metzler (ed.). Integration in the
early Roman West. The role of culture and ideology. Luxemburg: 47-64
— 2001. “The Lower Rhine triquetrum coinages and the ethnogenesis of the Batavi”, in: T.
Grünewald (ed., with H.-J. Schalles). Germania inferior. (RGA suppl. voi. 28). Berlin &
New York: 93-145
Rüger, C.B. 1998. “Lateinische Schriftlichkeit im römischen Grenzgebiet gegen die Germa¬
nen”. in: Düwel 1998: 357-375
Russell, P. 1995. An introduction to the Celtic languages. London & New York
Rybová, A. & P. Drda 1994. Hradiste by Stradonice. rebirth of a Celtic oppidum. Prague
Sakar, V. 1965. “Znacky na bronzovÿch nádobách starsi doby rimské v cechách”. Památky
Archeologické 56: 149-160
Salac, V. 1992. “Die Aussagen der Keramik zu Kontakten zwischen Böhmen und Mittel¬
deutschland in der Latènezeit”. in: Kelt.-germ. Besiedlung: 69-112
— 1997. “Vÿznam Labe pro cesko-saské kontakty v dobë laténské (úvod do problematiky)”.
Archeologické rozhledy 49: 462-494
— 1998. “Die Bedeutung der Elbe”. Germania 76: 573-617
Schindler-Kaudelka, E., & S. Zabehlicky-Scheffenegger 1995. “Die bodenständige Keramik
vom Magdalensberg. Ein Anfang”, in: Kelten, Germanen, Römer im Mitteldonaugebiet vom
Ausklang der Latène-Zivilisation bis zum 2. Jahrhundert. Brno - Nitra: 177-198
Schmidt, K.H. 1980. “Continental Celtic as an aid to the reconstruction of Proto-Celtic”. Zeit¬
schrift für vergleichende Sprachforschung 94: 172-197
— 1999. “Galatische Sprachreste”. in: E. Schwertheim (ed.). Forschungen in Galatien. (AsiaMinor-Studien 12). Ronn: 15—28
Scholz, M. 1997. “Namen von Kelten, ‘Römern’ und Germanen? Die Revölkerung von NidaHeddemheim im Spiegel von Namensgraffiti”, in: Bridger & Camap-Bomheim 1997: 4957
Schubert, F., 2002. « Ein neues Schriftzeugnis aus dem Oppidum von Manching », in : Sam¬
melblatt des Hist. Vereins Ingolstadt 110, 2001 (FS. S. Hofmann) : 37-55
Schulz, V. 1885/86. “Hrob u Zlivi”. Památky Archeologické 13
Schulz, W. 1939. Vor-und Frühgeschichte Mitteldeutschlands. Halle
Schumacher, S. 1992. Die rätischen Inschriften. Geschichte und heutiger Stand der Forschung.
(Innsbrucker Beiträge zur Kulturwissenschaft 79). Innsbruck
— 1993. “Das Etruskische und die ‘rätischen’ Inschriften aus der Sicht der Sprachwissen¬
schaft und Epigraphik”. Helvetia archaeologica 24: 31-50
— 1998. “Die ‘rätischen’ Inschriften: gegenwärtiger Forschungsstand, spezifische Probleme
und Zukunftsaussichten”, in: Ciurletti & Marzatico 1998: 334—369
Seebold, E. 1991. “Die Herkunft der Runenschrift”, in: J.O. Askedal, H. Bjorvand, E.F. Halvorsen (eds.). Festskrift til Ottar Gr0nvik pa 75-ársdagen. Oslo: 16—29 f
— 1993. “FuJ)ark, Beith-Luis-Nion, Ha-Lamedh, Abgad und Alphabet. Über die Systematik
der Zeichenaufzählung bei Buchstabenschriften”, in: F. Heidermanns, H. Rix, E. Seebold
(eds.). Sprachen und Schriften des antiken Mittelmeerraums. Fs J. Untermann. Innsbruck
Seidel, M. 1999. “Siedlungsfunde der Przeworsk-Kultur aus Hanau-Mittelbuchen, Main-Kinzig-Kreis (Hessen)”. Alt-Thüringen 33: 181-230
Sinnhuber, K. 1949. Die Altertümer vom Himmelreich bei Wattens: Ein Beitrag zur Vor¬
geschichte des tirol. Unterinntales (Schlem-Schriften 60). Innsbruck
Solinas, P. 1994. “Il celtico in Italia”. Studi Etruschi 60 (1995): 311—408
— 1999. “Su due iscrizioni in alfabeto leponzio provenienti dal Veronese”. Quaderni di arch¬
eologia del Veneto 15: 61-65
Stifter, J. 1989. “[Fundchronik] Strebersdorf”. Fundberichte aus Österreich 27 (1988): 301-02
Stoklund, M., 1995. “Die Runen der römischen Kaiserzeit”, in: (ed.) U. Lund Hansen et al.
Himling0je - Seeland — Europa, (Nordiske Fortidsminder B 13). Copenhagen: 317-346
— 1998. “Neue Runenfunde aus Skandinavien. Bemerkungen zur methodologischen Praxis,
Deutung und Einordnung”, in: Düwel 1998: 55-65
Stöllner, T. 1999. Der prähistorische Salzbergbau am Dürrnberg bei Hallein I: Forschungs¬
geschichte - Forschungsstand - Forschungsanliegen. (Dürmberg-Forschungen 1). Rahden/
Struck, M. 1996. Römische Grabfunde und Siedlungen im Isartal bei Ergolding, Lkr. Landshut.
Tejral, J. et al. (eds.) 1995. Kelten, Germanen, Römer im Mitteldonaugebiet vom Ausklang der
Latène-Zivilisation bis zum 2. Jahrhundert. Bmo-Nitra
Tibiletti Bruno, M.G. 1990. “Nuove iscrizioni camune”. Quaderni Comuni 49/50: 29-171
— 1992. “Gli alfabetari”. Quaderni Comuni 60: 309-380
Turetschek, F. 1976. in: Fundberichte aus Österreich 14 (1975): 129-130
Uenze, H.P. 1993. “Symbolgut”, in: Keltjt .: 189-192
— 2000. “Die jüngerlatènezeitliche Siedlung von Egglfing”. Bayerische Vorgeschichtsblätter
65: 1-38
Uslar, R.von 1991. Vorgeschichtliche Fundkarten der Alpen. (Römisch-Germanische Forsch¬
ungen 48). Mainz
— 1996. “Zu Rätern und Kelten in den mittleren Alpen”. Bericht der Römisch-Germanischen
Kommission 77: 155—213
Venclová, N. 1973. “Otázky etnické pnlusnosti podmokelské a kobylské skupiny”. Archeologické rozhledy 25: 41—71
— 1988. “Das Grenzgebiet der Latènekultur in Nordwestböhmen”, in: F. Horst & F. Schiette
(eds.). Frühe Völker in Mitteleuropa. Berlin: 121-127
Verger, S. 1998. “Note sur un graffito archaïque provenant de l’habitat hallstattien de Montmorot (Jura)”,
1998. Paris:
in: Académie
619—632des Inscriptions et Belles-Lettres, comptes rendus des séances de
Vitali, D. & G. Kaenel 2000. “Un Helvète chez les Etrusques vers 300 av. J.-C.”. Archäologie
der Schweiz / Archéologie suisse 23/3: 115—122
Völling, T. “Frühgermanische Expansion und einheimische Bevölkerung — Überlegungen am
Beispiel des großromstedtzeitlichen Gräberfeldes von Aubstadt im Grabfeldgau”, in: Kelt.germ. Besiedlung: 153—161
— 1995. Frühgermanische Gräber von Aubstadt im Grabfeldgau (Unterfranken). (Materialhefte
zur bayerischen Vorgeschichte A 67). Kallmünz/Opf.
Waldhauser, J. 1992. “Keltische Distributionssysteme von Graphittonkeramik und die Aus¬
beutung der Graphitlagerstätten während der fortgeschrittenen Latènezeit. Archäologisches
Korrespondenzblatt 22: 377-392
— 1992a. “Zum Bevölkerungswechsel von Kelten und Elbgermanen in Böhmen nach Sied¬
lungsstrukturveränderungen in der Mikro-, Mezzo-und Makroebene während der Endlatène-und frühen Kaiserzeit”, in: Kelt.-germ. Besiedlung: 162-183
— 1996. “Regionale keramische Kreise der jüngeren Latènezeit in Böhmen: Anfänge der For¬
schungen”. in: Jerem et al. 1996: 335-357
Waldhauser, J. et al. 1993. Die hallstatt-und latènezeitliche Siedlung mit Gräberfeld bei Radovesice in Böhmen. (Archeologickÿ vyzkurn v sevemích cechách 21). Prague
Wigg, A. 1999. “Confrontation and interaction: Celts, Germans and Romans in the central Ger¬
man highlands”, in: Creighton & Wilson 1999: 35-53
— 1997. “Germanen und Römer im Gießener Lahntal von augustäischer Zeit bis zum 3. Jahr¬
hundert”. in: Bridger & Camap-Bomheim 1997: 59-65
Wigg, D.G. 1999. “The development of the monetary economy in N Gaul in the late La Tène
and early Roman periods”, in: Creighton & Wilson 1999: 99-124
Wigg, D.G. & J. Riederer 1998. “Die Chronologie der keltischen Münzprägung am Mittel¬
rhein”. in: U. Peter (ed.). Stephanos Nomismatikos. Edith Schönert-Geiss zum 65. Geburts¬
tag. Berlin: 661-674
Williams, H. 1996. “The origin of the runes finally revealed”. Amsterdamer Beiträge zur älteren
Germanistik 45: 211-218
— 1997. “The Romans and the runes — uses of writing in Germania”, in: S. Nyström (ed.).
ochABC. Elvaföreläsningar fr&n ett symposium i Stockholm váren 1995. Stockholm:
Willvonseder, K. 1932. “Neue Latènefunde aus Niederösterreich”. Germania 16: 272-275
Wyss, R. 1954. “Das Schwert des Korisios. Zur Entdeckung einer griechischen Inschrift”,
Jahrbuch des Bernischen Historischen Museums in Bern 34 (Bern 1955): 201-221
Zabehlicky, H. 1991. “Ritzungen auf Grabbeigaben aus Hallstatt”. Specimina Nova Diss. Inst.
Hist.(Pécs): 271-280
Zabehlicky-Scheffenegger, S. 1997. “Dreifußschüsseln mit Töpfermarken vom Magdalens¬
berg”. Rei Cretariae Romanae Fautorum Acta 35: 127—132
— & R. Sauer, 1998. “Olle ad impasto di grafite del Magdalensberg: forme, funzione ed analisi
archeometriche”, in: S. Santoro Bianchi & B. Fabbri (eds.), Il contributo delle analisi
archeometriche allo studio delle ceramiche grezze e comuni, il rapporto forma/funzione/
impasto , Bologna 1997 (pubi. 1998), 87-91 (English summary p. 265)
Zeller, K.W. 1989. “Neue keltische Gewerbebauten auf dem Dürmberg”. Salzburger Archiv 6:
Ziegaus, B. 1995. Der Münzfimd von Großbissendorf. Eine numismatisch-historische Untersuch¬
ung zu den spätkeltischen Goldprägungen in Südbayern. Munich
— 1997. “Datierung boischer Münzen durch die Analyse von Schatzfunden” = chapter
6.4.5. 1.1.1 in: Das prähistorische Gold in Bayern, Böhmen und Mähren: Herkunft - Tech¬
nologie — Funde, vol. I: Textband. (Památky archeologické suppi. 7): 213-221
— 2000. “Der Goldfund”, in: S. Sievers et al., “Vorbericht über die Ausgrabungen 1998-1999
im Oppidum von Manching”. Germania 78/2: 378—382 (complete article p. 355-394)
Ziegaus, B. & H. Rix 1998. “Ungewöhnliche Funde der späten Hallstattzeit aus dem Vor¬
alpenland”. Germania 76: 291-303
(including marks on metal and wood; in [ ]: number of occurences except on coins, total: 665
(in addition: 582 marks on coarse grey ware, sum total 1247)
Al* — , A2* — , A3* •-•on coins only; Roymans 2001: 102 f.; Forrer 1910: 455.
Bl I [>7]
Katsch, Styria, tomb II: Ehrenreich 1994: 16 f; 34 pi. 5,1. — Magdalensberg: ZabehlickyScheffenegger 1997: 131. — Manching, PSS inv.no. 1974, 1772 [+ 2x D2]: Krämer 1982: 497
fig. 5,1. — Stare Hradisko [2]: Cizmár & Meduna 1985: 97 f. — Roseldorf (?): Maurer 1980:
411. 413 fig. 406.
Wood: Hallstatt, Christinastollen [2]: Barth 1984.
B2 Ψ [3]
Stare Hradisko [3]: Cizmár & Meduna 1985: 97 f.
B3 [1]
Stare Hradisko: Cizmár & Meduna 1985: 97 f.
B4 II [2]
Karlstein, Lkr. Berchtesgaden; Neubau near Linz (or E3+B1?): Kappel 1969: 116; fig. 42,18;
B5 K [1]
Katsch, Styria, grave IX: Ehrenreich 1994: 35; pi. 6,1.
B6 3 [1]
Katsch, Styria, grave II: Ehrenreich 1994: 16 f; 34 pi. 5,1.
B7 III [>2]
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131. — Stare Hradisko [2]: Cizmár & Meduna 1985:97.
Coins: De la Tour, pi. 40, no. 9475.
B8* k, U on coins only
De la Tour, pi. 40, no. 9474.
B9 λ , Ί [2] on wood only
Hallstatt, Christinastollen [2]: Barth 1984.
BIO 1 [1]
Grub, Groß St. Florian: Hebert 1991: 283; 291 fig. 5.
Bll E,M,
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131.
Coins (E): De la Tour, pi. 40, no. 9474.
B12 K [2] on wood only
Hallstatt, Christinastollen [2]: Barth 1984.
B13 H-on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
B14 o B on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
B15 :i on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
Cl Ψ, A [8]
Katsch, Styria, tomb IX: Ehrenreich 1994: 35; pi. 6,1. — Linz Gründberg: Franz 1942: 45; pi.
35,3. — Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; Zabe¬
hlicky-Scheffenegger 1997: 131. — Praha 6 -Bubenec: Sakar 1965: 153 fig. 2,2 quoting Novo¬
tny 1948. — Stradonice Hradist, okr. Beroun; Trisov, okr. Cesky Krumlov: Kappel 1969: 189
Metal dishes: Schlabitz (Tschiläsen), Kr. Guhrau (Silesia): Jahn 1919: 21 fig. 3; Pescheck
1939: 135.
Wood: Hallstatt, Christinastollen [3]: Barth 1984.
Also often on coins (type Forrer 398): Kappel 1976: 88-90.
C2 Y, Y, T [67]
Boritov: Cizmár & Meduna 1985: 94. — Devin, okr. Bratislava-mesto: Kappel 1969: 189 f. —
Dobrichov, okr. Kolin [?]: unpubl. (N. Venclová, pers.comm.) — Drnholec [2]: Cizmár &
Meduna 1985: 94.— Dürnkrut, Bez. Gänserndorf: Hörler 1953: 56; 199 pl. 106; Kappel
1969: 189. — Frauenstein, Bez. Braunau a.Inn; Hallstatt Dammwiese [12; 1?]: Kappel 1969:
189 f. — Hallstatt, in the Lahn [2]: Hell 1949: 81-83. — Hrazany, okr. Pribram [> 3]; Karlstein, Lkr. Berchtesgaden: Kappel 1969: 188; 190. — Kelheim oppidum : Etzel 1990: 30; pl.
50,16. — Klentnice, okr. Bfeclav [2]: Kappel 1969: 190; Cizmäf & Meduna 1985: 95. — Leibenfeld, Deutschlandsberg: Hebert 1991: 282 f; 290 fig. 4. — Linz Freinberg; Linz Gründ¬
berg: Kappel 1969: 188—190. — Magdalensberg: Zabehlicky-Seheffenegger & Sauer 1998:
89 fig.6 (?); Zabehlicky-Seheffenegger 1997: 131 and pers. comm. — Manching [4]: Kappel
1969: 188—190; Franz 1942: 45; pl. 35,4 (Gründberg). — Manching, pit 816d: Maier 1992: pl.
106, 19. — Micheldorf, Bez. Kirchdorf a.d. Krems: Kappel 1969: 189. — Mistrin: Cizmär &
Meduna 1985: 96. — Musov, okr. Breclav: Cizmaf 1995: 114; 115,5.— Nejdek okr. Breclav;
Polesovice: Cizmäf & Meduna 1985: 97. — Ratschendorf (2nd c. AD): Artner, unpubl. (see
Hebert 1991: 286 fn. 28). — Salzburg Rainberg: Franz 1942: 45; pl. 35,6; Kappel 1969:
189. — Stare Hradisko [3; 1?]: Cizmäf & Meduna 1985: 97 f. — Stochov, okr. Kladno: Franz
1942: 45; pl. 35,7; Kappel 1969: 190.— Strachotín: Cizmäf & Meduna 1985: 99. — Stradonice Hradist, okr. Beroun [7]: Franz 1942: 45; pl. 35,7; Kappel 1969: 190; fig. 43. — Stramberk ‘Kotouc’: Cizmäf & Meduna 1985: 99. — Trisov, okr. Cesky Krumlov (prev. Hollubau) [>
6]: Franz 1942: 45; pl. 35,7; Kappel 1969: 190. — Wien III [2]; Wien, Bisamberg fr. l.Jh.n.:
Kappel 1969: 190; Wien III, Rudolfstiftung (?): Ehrenhöfer & Pichler 2001: 287 pl. 1,5. —
Wittenberg: Schulz 1939: 154 fig. 188.
C3 Y, Y [48]
Altendorf, Lkr. Bamberg: Pescheck 1976: 369; 370 fig. 2. — Kelheim; Manching [27; 1?];
Hallstatt Dammwiese [12; 2?]; Neubau, Bez. Linz-Land: Kappel 1969: 188 f; fig. 29,9. —
Salzburg, Heilbrunner Berg: Franz 1942: 45; pl. 35,5; Kappel 1969: 189. — Stradonice Hra¬
dist, okr. Beroun (?); Trisov, okr. Cesky Krumlov [2]: Kappel 1969: 190.
C4 Y [12]
Hallstatt, Dammwiese [4]: Franz 1942: 45; pl. 35,2; Kappel 1969: 189. — Manching [6];
Straubing: Kappel 1969: 188 f.; fig. 27, 8.
Doubtful: Dürnkrut, Bez. Gänserndorf (LT D, type C4-6): Willvonseder 1932: 275; Kappel
1969: 189.
C5 Y [2]
Hallstatt Dammwiese; Manching: Kappel 1969: 188 f
C6 X [1]
Ratschendorf, Hügelstaudach, mound 17: Artner 1995: 39; 64 pl. 4, no. 2
C7 Y [7]
Hallstatt Dammwiese [5; 1?]; ?); Trisov, okr. Cesky Krumlov: Kappel 1969: 189 f
C8 T [3]
Hallstatt Dammwiese [1; 2?]: Kappel 1969: 189.
C9 T [2]
Altendorf (Lkr. Bamberg): Pescheck 1976: 369. 370 fig.l. — Hallstatt, Dammwiese: Barth &
Leutner 1993: 462. fig.543.
CIO Y[l]
Stare Hradisko: Cizmár & Meduna 1985: 97 f.
Cll '/ [1]
Katsch, Styria, grave IX: Ehrenreich 1994: 35; pi. 6,1.
C12 Y [1]
Waltersdorf an der March: Turetschek 1976: 130; fig. 203
C13 Y on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
C15 Y on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
D1 Λ, Λ, V [>6]
Ladendorf (frgm.): Maurer & Strouhal 1982: 440. 441 fig. 462. — Lysá nad Labem, okr.
Melnik: Sakar 1965: 157 fig. 5 (map). — Magdalensberg: Zabehlicky-Scheffenegger 1997:
131. — Oberthiirnau: Bauer & Maurer 1978: 380; 382 fig. 214. — Pistorf: Hebert 1991:
283; 292 fig. 7. — Praha 6-Bubenec (“1948”); Trebusice, okr. Kladno: Sakar 1965: 157 fig.
Also often on coins (type Forrer 398): Kappel 1976: 88—90.
D2 X, X, X [21]
Berchmg-Pollanten, excavation 1986 (unpubl.: M. Leicht, pers. comm.). — BudapestGellérthegy-Tabán (possibly D5): Kappel 1969: 191. — Eszterg om, Komárom m. (double
line): Kappel 1969: 191. — Iza: Kappel 1969: 116. — Katsch, Styria, grave IX: Ehrenreich
1994: 35; pl. 6,1. — Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995:
198; Zabehlicky-Scheffenegger 1997: 131. — Manching, PSS inv.no. 1974, 1772 [2, +B1]:
Krämer 1982: 497 fig. 5,1.— Manching [2], PSS inv. nos. 1974, 1315; 1974, 2182: Krämer
1996: 368 fig. 5,4; 5,7.— Némcicky: Cizmár & Meduna 1985: 97. — Oslavany, okr. Bmovenkov [2]: Cizmár 1995: 114, 1-2. — Praha 6-Bubenec [2]: Sakar 1965: 153 fig. 2,6. —
Stare Hradisko [5]: Cizmár & Meduna 1985: 97 f. — Wien, Leopoldsberg: Adler & Schön
1979: 306. 305 fig. 265.
Doubtful: Hallstatt: Kappel 1969: 116; fig. 37,15.
Also on coins: Ziegaus 2000.
D3 X [1]
Stare Hradisko: Cizmár & Meduna 1985: 97 f.
D4 * [1?]
Strebersdorf (thin incision on samian ware, doubtful): Stifter 1989: 301. fig. 436
D5 + or -h [>1]
(D5 ist not grouped in section B because it often cannot be distinguished from D2)
Großrußbach, Niederösterreich: Adler & Siegmeth 1988: 225. fig. 349. — Magdalensberg:
Zabehlicky-Scheffenegger 1997: 131.
D6* /\ [1]
Praha 6-Bubenec: Sakai 1965: 153 fig. 2,4.
Coins (type Forrer 398): Kappel 1976: 88—90.
D7 A [2]
Metal dish: Zbv, okr. Jicin: Schulz 1885/86: 80; Sakar 1965: 159.
Doubtful: Berching-Pollanten inv. no. 1984/ 154: Lasch 1999: 49; 242; pi. 83,24.
D8 X [1]
Berching-Pollanten inv. no. 1984/4: Lasch 1999: 49; 232; pi. 57,35.
D9 X, »<, X [>2]
Magdalensberg: Zabehlicky-Scheffenegger & Sauer 1998: 89 fig.6; Zabehlicky-Scheff¬
enegger 1997: 131 and pers.comm.
Wood: Hallstatt: Barth 1984.
DIO* A, Dll* A, D12* < on coins only
Roymans 2001: 102 f; Kappel 1976: 88-90; de la Tour, pi. 40, no. 9475.
D13 Â on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
D14 / on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
D15 II,//
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131 and pers.comm.
Coins (type Forrer 398): Kappel 1976: 88-90.
D16 # on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198.
El « [1]
Karlstein, Lkr. Berchtesgaden: Kappel 1969: 188; pi. 31,9.
E2 H,
(?) [16]
Drnholec; Klentnice; Milovice (13x); Stare Hradisko: Kappel 1969: 116; Cizmár &
Meduna 1985: 97 f.
E3 ΛΛΛ, --- [2],
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131 and pers. comm.; Zabehlicky-Scheffenegger & Sauer 1998: 89
fig.6. —Wien (above D2) [1; 1?]: Kappel 1969: 116; fig. 48.2-3.
Coins : e.g. Leicht & Ziegaus 2000; Ziegaus 2000.
E4* V [1]
Metal dish: Zliv, okr. Jicin: Schulz 1885/86: 80; Sakar 1965: 159.
E5 4 [1]
Wood: Hallstatt: Barth 1984.
E6* \Ä/ on coins only
Ziegaus 2000.
E7 zzz, Icsd on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131.
E8 >,|f/ on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131 and pers. comm.; Zabehlicky-Scheffenegger & Sauer 1998: 89 fig.6.
E9 O 3 on coarse grey ware only
Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198.
(?) on coarse grey ware only
Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198.
Ell° lit on coarse grey ware only
Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198.
Fl N, X [3]
Stare Hradisko [2]: Cizmár & Meduna 1985: 97 f.
Doubtful: Manching (jy"j): Krämer 1982: 498, fig. 6.
F2 H [1]
Katsch, Styria, grave IX: Ehrenreich 1994: 35; pi. 6,1.
F3* □ on coins only
De la Tour, pi. 40, no. 9475.
F4* O, £ (?) on coins only
De la Tour, pi. 40, no. 9474.
F5*> on coins only
De la Tour, pi. 40, no. 9475.
F6 ™ on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
F 7 O |M (?) on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
F8 ìo7 (?) on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198 ZabehlickyScheffenegger 1997: 131.
Gl X,X (?),I [3]
Berching-Pollanten, excavation 1986 (archs overlapping, engraved over HI; unpubl.: M.
Leicht pers. comm.). — Manching: Kappel 1969: 188; pi. 25,679 (doubtful?)
Doubtful : Kelheim, oppidum,·. Leicht 2000: 97 (no trace of middle line); pi. 17,3.
G2 X [1]
Devin, okr. Bratislava-mesto: Kappel 1969: 190; 102 fig. 48,7.
G3 X [7]
Manching (3; 4?): Kappel 1969: 188; pi. 25, nos. 676; 677; 678; 680; 681-682.
G4 Φ [3?]
Egglfing, Köfering, Lkr. Regensburg (double circle): Uenze 2000: 34, Abb. 14, 5. — Linz,
Römerberg: Kappel 1969: 189; 97 fig. 42,16 (small sign on the rim). — Wien (?): Ehrenhöfer &
Pichler 2001: 287 pi. 1, 5.
G5 t> [3]
Manching: Kappel 1969: pi. 26, 694 (?); 698; 700.
G6 e [3]
Hallstatt Dammwiese (3x, 2x ?): Kappel 1969: 189; 93 fig. 37,14.
Coarse grey ware: Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
Coins, doubtful : Forrer 1910: 454 f.
G8 C on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
HI O, o, -Ò-[4]
Berching-Pollanten, excavation 1986 (engraved under Gl; unpubl.: M. Leicht
pers.comm.). — Magdalensberg: Sehindler-Kaudelka & Zabehlicky-Scheffenegger 1995:
198. — Manching: Kappel 1969: pi. 26, no. 705. — Nosislav: Cizmár & Meduna 1985: 97.
Doubtful: Kelheim oppidum·. Etzel 1990: 59; pi. 51,17.
H2 ®, © [15]
Dürrnberg: Kappel 1969: 116. — Manching [9]: Kappel 1969: pl. 26, nos. 693; 695; 696;
699; 701b; 702b; 704 (?); 706; 707,— Mistrin: Kappel 1969: 116; Nejdek [2?]: Kappel 1969:
116; Cizmár & Meduna 1985: 96. — Lovcicky, okr. Vyskov: Cizmár & Meduna 1985: 95. —
Straßkirchen: Kappel 1969: 116.
H3 Θ, © [(21)]
Postoupky (series of 21 small signs; stamped): Cizmár & Meduna 1985: 97.
H4 © [4]
Mistrín; Stare Hradisko [2 stamped, 1?): Cizmár & Meduna 1985: 96-98.
H5 Θ, $, ©, ® [12]
Berching-Pollanten, inv.no. 1984/4: Lasch 1999: 49; 232; pl. 57,35. — Manching: Kappel
1969: 116 no. 689; pi. 37, nos. 871b; 879; 880; Pingel 1971: pl. 97,1517.— Stuttgart Stamm¬
heim: Fundberichte aus Schwaben NF 11 (1938—50), 97; Kappel 1969: 119 fn. 230.
Doubtful: Berching-Pollanten, inv.no. 1984 /279: Lasch 1999: 49; 242; pl. 83, 23. — Man¬
ching, pit 816d: Maier 1992: pl. 110,12. — Manching, PSS inv. no. 1974, 1772: Krämer 1996:
368 fig. 5,3. — Pfaffenberg: Ehrenreich 1994: 738, fig. 575 (Roman). — Wien XXIII: Hahnei
1994: 56 fig. 36 (Roman).
H6 © [H5 or H6: 8]
Manching: see Kappel 1969: 112; 116
H7 Ö[l]
Manching: see Kappel 1969: 112; 116; pl. 26, no. 703.
H8 è (?) [1]
Manching: Krämer 1996: 368, fig. 5,5.
H9 * Φ, Φ on coins only
Roymans 2001: 102 f; Forrer 1910: 454 f.
H10* © doubtful; on coins only
Forrer 1910: 454 f.
Hll O,
on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198; ZabehlickyScheffenegger 1997: 131.
II ·
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
Coins: Roymans 2001: 102 f.
12* ·*·, 13* :·:, 14* %.· on coins only: Roymans 2001: 102 f.
15 ; on coarse grey ware only
Magdalensberg: Schindler-Kaudelka & Zabehlicky-Scheffenegger 1995: 198.
16 ·». on coarse grey ware only
Magdalensberg: Zabehlicky-Scheffenegger 1997: 131.
Fragmentary, not identified [280]
Au, Lkr. Laufen: Kappel 1969: 188-190. — Berching-Pollanten inv.no. 1983/ 732 (type H):
Lasch 1999: 49; 239; pi. 77,20. — Biberg, Saalfelden, Bez. Zell am See: Kappel 1969: 189
f. — Boritov, okr. Blansko; Brest, okr. Kromeriz; Bykovice, okr. Blansko: Cizmár 1995: 109;
112,2-3; 113, 9. — Devin, okr. Bratislava-mesto: Kappel 1969: 188-190. — Dietmannsdorf:
Hebert 1991: 283; 292 fig. 6. — Dürrnberg, Bez. Hallein [7; 1?]; Hallein (?); Hallstatt
Dammwiese [111 frg.]: Kappel 1969: 188-190. — Hostÿn, Chvalcov, okr. Kromeriz: Cizmár
1995: 109; 113,1; 3; 5; 10. — Hrazany, okr. Pribram: Kappel 1969: 189 f. — Hulin, okr.
Kromeriz; Jicina, okr. Novy Jicin [2]: Cizmár 1995: 109; 113,1; 3; 5; 10.— Karlstein, Lkr.
Berchtesgaden [4; 2 type D?]: Kappel 1969: 188—190. — Koprivnice, okr. Novy Jicin;
Laskov, okr. Prostëjov [2]; Lanzhot. okr. Breclav [3]: Cizmár 1995: 109; 113,2; 8; 11; 115,6;
9;10.— Linz Freinberg [7]; Manching [79]: Kappel 1969: 188-190. — Manching, pit 795b
[2]; ditch 852-1; pit complex 791; pit 815a [2]; pit 830a/b [5]: Maier 1992: pis. 45,3-4; 46,13;
65,16; 88,8 & 93,6; 131,1—4 & 143,11. — Mistrín, okr. Hodonín; Moravské Kninice, okr.
Bmo-venkov; Musov, okr. Breclav: Cizmár 1995: 109; 112,4; 115, 7;8. — Neubau, Bez. LinzLand (“some fragments”); Nitriansky Hrádok, okr. Nové Zámky [2]: Kappel 1969: 189 f—
Praha 6 Bubenec: Sakar 1965: 153 fig. 2,6; different item (j Hahnentritt ): 157 fig. 5 (map). —
Salzburg, Rainberg; Straubing [2]: Kappel 1969: 189 f. — Stradonice, okr. Beroun, Hradist
[2]: Kappel 1969: 189 f. — Trisov, okr. Ceskÿ Krumlov [21]: Kappel 1969: 189 f. — Yolders/
Wattens, Bez. Innsbruck; Wien, Leopoldsberg [2],
Doubtful: Salzburg Maxglan: Franz 1942: 45 n.2 (cf. Kappel 1969: 189 f)·
Privat Dozent Dr Jürgen ZEIDLER
Universität Trier
Fachbereich III — UB 31
D-54286 Trier
e-mail <zeidler@uni-trier.de>