Eburo ?

[Matrix]

Bj le Grand Alexandre

Bref, tout est négat-if ?

tu as dis

"Le latin ebrius résulte d'une variante h2eghw-r-y-os sur la racine h2ekw- illustrée par le hittite ekuzi = boire, directement apparenté à l'un des noms indo-européens de l'eau "

Quid alors du Latin Abr-umus = Sevré

[Jacques]

Quid alors du Latin Abr-umus = Sevré

Je pressens la coupure ab-rumus (ab préfixe privatif)

[Alexandre]

Le mot paraît isolé. Je ne suis pas un grand latiniste, mais je pencherais vers un adjectif déverbal à partir d'une forme apparentée à ab-rumpo = quitter en rompant les liens. En tout état de cause, la remarque de Jacques me paraît fondée.

[Matrix]

Tout à fait Thiery, euh Jacques, heu Alexandre

Ab-rumus = sans mamelle (de ruma)
j'ai vérifié suite à tes indications

yavait aussi un nom Gaulois très attesté :
Ab-rara-courcix mais je pense que tu sais d'où il vient

[Dvrnacos]

Jacques, vous avez raison : le nom gaulois de Ebersmunster est Noviento, donc rien à voir avec Eburo.

Je voudrais juste signaler que les noms de lieux que que nous avons évoqués (Saint-Prix et Ebersmunster) ont été donnés à des périodes (676 après JC dans votre cas et 723 après JC dans mon cas) où l'évangélisation battait son plein. Et je pense, comme beaucoup sur ce forum, que ceci fait écho à une volonté délibérée de re-baptiser des lieux majeurs pour les croyances dites païennes.

De NOVIENTO à EBERSMUNSTER...
Au temps des Celtes, l'île de NOVIENTO (appellation primitve du lieu) formée par les bras de l'Ill, constitue un lieu de culte en l'honneur du Dieu païen TEUTATES. Au temps Gallo-Romains (1er Siècle avant JC - 3ème Siècle après JC), NOVIETUM est un lieu de culte en l'honneur des divinités romaines MERCURE et DIANE. Peut-être dès cette époque, une petite communauté de chrétiens s'établit à cet endroit.
Vers 675, l'abbé-évêque irlandais DEODAT groupe une communauté de moines sur le domaine donné par le Duc d'Alsace ADALRIC et son épouse BERESWINDE, parents de Sainte ODILE. DEODAT ne reste que peu de temps dans le monastère qu'il a fondé en Alsace. Il part à Saint Dié pour y fonder un autre monastère et lui lègue son nom. Après la migration des peuples, le monastère de EBROTHEIM (723 après JC) devient un centre de christianisation de la Moyenne Alsace.
La famille de Ducs d'Alsace fait de nombreuses donations à l'abbaye d'Ebersmunster.

La suite sur http://mairie.ebersmunster.free.fr/histo.htm

[Dvrnacos]

La piste Eburo-nom d'arbre semble être aussi conservée outre-Rhin : en allemand, Eberesche ou Vogelbeere (Sorbus aucuparia) correspond, en effet, au sorbier des oiseleurs.

[Matrix]

les éburons = les cultivateurs d'ifs

ici

http://fr.wikipedia.org/wiki/%C3%89burons

[Jacques]

Un florilège d'affirmations non fondées, de raisonnements inversés ; d'ailleurs le préambule de wikipédia prévient que les auteurs ne citent pas de sources.

[Matrix]

Merci Jacques
J'avais mis le lien..........sans avis
car je le sentais pas non plus

[Matrix]

Bonjour
je ressuscite un vieux topic concernant l'EBURO Celtique = if et l'EBURO Germanique = sanglier
cf pièce jointe et j'essaye de faire un copié/collé
l'auteur penche pour un EBURO = if avec quelques nuances sur le végétal désigné


mélanges en l’honneur de pierre-yves lambert 65
The meaning of Celtic *eburos
1. Introduction
There is no doubt that Proto-Celtic possessed a phytonym *eburos. It survives as Old Irish ibar,
Middle Welsh efwr, Middle Breton (h)evor. Although we lack control over their lexical meanings,
numerous Continental Celtic names beginning with Ebur(o)- can be connected with this etymon,
too. The general assumption is that the original meaning of the phytonym is ‘yew tree’: Sanz et
al (2011, 450-1), Matasović (2009, 112), Sims-Williams (2006, 78) and Delamarre (2003, 159-
60) are some of the most recent proponents of that idea. A notable exception is Dagmar Wodtko
(2000), who did not assign a meaning to the proto-form. The aim of this paper is to show that
*eburos did not mean ‘yew tree’.
2. The evidence for *eburos meaning ’yew’
The only strong argument in favour of the meaning ‘yew’ is the well attested meaning of the
Irish word: OIr. ibar, MoIr. and Sc.G. iubhar ‘yew’ (Taxus baccata). Welsh efwr, however, denotes
‘hogweed, cow parsnip’ (Heracleum sphondylium), a large perennial umbellifer, while Breton evor
denotes ‘buckthorn’ (Frangula alnus), a woody shrub. There is very little that these three plants
have in common botanically. It is not at all clear why we should single out Irish and select the
meaning ‘yew’ as the Proto-Celtic meaning.
The Continental Celtic material provides plentiful formal evidence for the presence of *eburoin
personal names, place-names and, due to a recent find from Spain, a theonym.1 As usual in
such cases, however, the semantic information that is provided by names is non-existent at worst
and inferential at best. A case in point is the name Eburones of a people that lived in present-day
northern Belgium and the southern part of the Netherlands. There is a famous story reported by
Caesar which states that after their crushing defeat against Caesar one of the two leaders of the
Eburones, Catuvolcus, poisoned himself by ingesting yew.
Catuvolcus ... taxo, cuius magna in Gallia Germaniaque copia est, se exanimavit
‘Catuvolcus killed himself by yew, which is widespread in Gallia and Germania’ (Caesar,
De Bello Gallico VI.31)
This has been taken as an indication that the Eburones had a special connection with the yew,
1 Sanz et al 2011.
Peter Schrijver
University of Utrecht
66 mélanges en l’honneur de pierre-yves lambert
which was formalized in their name.2 If Eburones meant ‘yew people’, that would certainly make
for a nice story, but it is worth stressing that this particular point apparently escaped the storyteller,
Caesar, who does not mention the connection explicitly. Furthermore, since the pharmacological
properties of yew have presumably always been well known, using it as a poison cannot have been
the prerogative of a people whose name contained a word meaning ‘yew’.
Another indication that in Continental Celtic *eburos may have denoted the yew is derived
from the name of the Aulerci Eburovices, a tribe of Central Gaul. Eburovices contains the root
*wik- ‘to fight’ as its second member. A similar compound, Lemovices, which is the tribe that gave
its name to present-day Limoges, literally means ‘Elm-fighters’, where ‘elm’ is not so much the
logical object of ‘fight’ as the logical instrument, referring to weapons made out of elm wood. So
Lemo-vices means ‘those who fight with elm’. Compare also the south Walian Ordo-vices ‘those who
fight with hammers’ of the Roman period. In this light it seems quite plausible that Eburovices
means ‘those who fight with yew’, i.e. with weapons such as bows and spears made out of yew.3
It is difficult to accept anything other than plausibility, however, since the semantic range of
words that can be combined with *wik- is relatively large, cf. Gabrantowikōn (a Greek genitive
plural, attested for Britain in Ptolemy’s Geographia) ‘those who fight like (he-)goats (*gabros)’,
Ogam Irish LUGUVVE CCA < genitive *lugu-wik-os ‘who fights like the god Lugus or like a lynx’,
RITUVVECAS < genitive *ritu-wik-os ‘who fights at a ford’. While one may readily accept that
*eburo-wik- means something like ‘who fights like or with (weapons made of ) the *eburos plant’,
the suggestion that *eburo- in Eburovices means ‘yew’, as in Irish, is no more justified than that it
means, say, ‘buckthorn’, as in Breton. The tough but elastic wood of buckthorn is well suited for
making arrows. Given the well-known laxative properties of buckthorn bark and berries, one might
for a moment even entertain the amusing thought that a person who ‘fights like a buckthorn’ was
eminent at instilling fear in his enemy. The idea that Gaulish *eburos meant ‘buckthorn’ rather than
‘yew’ receives potential support from Gamillscheg’s proposal that the French word for ‘buckthorn’,
bourdaine, Old French borjaine can be reconstructed as Late Latin *ebur-j-ēnā, which ultimately
derives from Gaulish *eburo-.4 The unclear word-formation as well as the early loss of initial *e- give
room for doubt about this etymology, however.
3. Is there a Celtic *eburos ‘wild boar’?
Pierre-Yves Lambert5 has looked favourably on the old idea6 that Eburovices rather than ‘yew’
contained a word for ’boar’, which is better attested in Germanic (e.g. Old English eofor, Old
High German ebur < *eburaz). In that case Eburo-vices means ‘those who fight like boars’. Lambert
supports that idea by referring to the fact that two coins of the Aulerci Eburovices show a picture
of a boar.7 While this is an attractive possibility,8 it is not entirely convincing: pictures of boars
2 E.g. Evans 1967, 347.
3 Delamarre 2003, 199 s.v. lemo-, limo-.
4 Gamillscheg 1969, 136.
5 Lambert 1994, 34.
6 E.g. Jullian 1920-1926, 35.
7 RIG IV, numbers 65 and 151.
8 See also Sanz et al 2011, 451-452.
mélanges en l’honneur de pierre-yves lambert 67
are so widespread in the ancient Celtic and Germanic world that their appearance on coins of the
Eburovices may well be unrelated to their name.
The idea that Continental Celtic names that contain *eburo- ultimately refer to boars rather
than plants is often referred back to Joseph Vendryes.9 Vendryes, however, spelled out rather than
embraced that possibility. He reacted to an article by Paul Courteault, who discussed a Latin
inscription on an altar found in Bordeaux in 1921.10 Given its impact on the discussion about
the meaning and etymology of *eburos, it is useful to discuss this altar in some detail. Courteault
presents the following reading of the inscription, which is based on autopsy by Camille Jullian:
DEAE TVTEL(a)E BOVDIG (ae)
M (arcus) AVR (elius) LVNARIS IIIIII
VIR AVG (ustalis) COL(oniarum) EBOR(aci) ET
LIND (i) PROV (inciae) BRIT(anniae) INF(erioris)
ARAM QVAM VOVER (at)
AB EBORACI AVE CT(us)
V (otum) S(olvit) L(ibens) M(erito)
PERPETVO ET CORNE(liano consulibus)
The reference to the consuls Perpetuus and Cornelianus dates the inscription to 237 AD . The
dedicant is the Roman citizen Marcus Aurelius Lunaris, who presents himself as sevir augustalis,
a member of a college charged with the cult of the Emperor, and as an inhabitant of the Roman
coloniae of York (Eboracum = Eburacum) and Lincoln (Lindum) in the province of Britannia Inferior.
The dedication is to the goddess Tutela Boudiga. Given the fact that the dedicant personifies the
link between Roman Britain and the important trading port Bordeaux, Tutela is no doubt Tutela
navis, the divine protector of ships and shipping. She is identified here with Boudiga, the Celticlanguage
equivalent to Victoria, the goddess that personifies Imperial rule and the well-being of the
Emperor, an issue close to the heart of this sevir augustalis.
The Latin of the inscription contains at least one curiosity: ab Eboraci instead of ab Eboraco
‘from Eboracum’, which Courteault took as a possible locative11 but which, given the relatively late
date of the inscription, may rather be a first sign of the Late Latin loss of the genitive, which led
to its hypercorrect use here. The text can be translated as follows: ‘To the goddess Tutela Boudiga,
Marcus Aurelius Lunaris, sevir augustalis of the coloniae of York and Lincoln of the province of
Lower Britannia, (has given) an altar, which he had pledged when he departed (avect(us)) from
York. He has made good his promise, of his own free will and because she deserves it’.
So far, the altar provides nothing that is relevant to determining the meaning of Celtic *eburos,
because all that the inscription gives us is the name of York, in the forms Ebor and Eboraci.
The importance of the altar resides in the way in which the provenance (York) and destination
(Bordeaux) of the dedicant link up with the sculptured reliefs on either side of the altar. The side
panel to the right shows a sitting half-naked horned deity holding an anchor in one hand and
resting his other hand on an urn from which water flows. This is the divine river Garunna, the
9 Vendryes 1923, 476-477.
10 Courteault 1922.
11 Courteault 1922, 240.
68 mélanges en l’honneur de pierre-yves lambert
Garonne, which symbolizes the town of Burdigala, Bordeaux. The side panel to the left shows a
wild boar standing on a socle. As Courteault argues, the symbolism of the altar suggests that the
boar represents Eburacum, York, which makes sense since the name Eburacum itself is associated
with Celtic *eburos meaning ‘wild boar’. For the existence of that word in Celtic, he refers to
Jullian.12 However, this is just one of the possibilities entertained by Jullian, who explains Eburovices
as ‘warriors or descendants’ of either ‘a divine Eburos’, or ‘wild boar’, or ‘yew’. Jullian (ibidem)
translates the tribal name Eburones as ‘Wild Boars’ or ‘Yews’, interpreting the latter as ‘Murderers’,
presumably on account of the poisonousness of yew. Given the way in which he presents these
possibilities on the page, Jullian appears to prefer the interpretation of *eburos as ‘boar’ but he does
not explain why. However this may be, it is difficult to deny that the Bordeaux altar associates the
toponym Eburacum with a boar.
Now Vendryes takes issue with Courteault.13 He explains correctly that the association of
Eburacum with a boar, which is plausible, cannot be justified on etymological grounds because
a Celtic etymon *eburo- ‘boar’ (rather than ‘yew’) simply does not exist. Subsequently, Vendryes
explains that, if nonetheless one wishes to connect Celtic *eburos with the word for ‘boar’ in Latin
aper, Germanic *eburaz and Old Church Slavonic veprь, this comes at the price of accepting one
of the following two additional assumptions that are required to bridge the formal gap between
those forms:
(1) Assume a proto-form *epro-: *epro- regularly becomes Celtic *ebro- according to the rule that
Proto-Indo-European *p becomes *b before *r, l, as in *piplāseti > OIr. -ebla ‘will drive’, *piprāseti >
OIr. -ebra ‘will bestow’.14
(2) Compare Celtic *eburo- first and foremost with Germanic *ebura- < earlier Indo-European
*epuro- rather than with any of the other forms: this comparison will only work if the relation
between Celtic *eburo- and pre-Germanic *epuro- is the same as in Celtic *gabro- ‘goat’ and non-
Celtic *kapro- (in Old Icelandic hafr, Latin caper).
So much for Vendryes’ alleged support for the existence of a Celtic *eburos meaning ‘boar’.15 In
light of our present knowledge of Celtic historical phonology, however, neither scenario that is
supposed to salvage the connection of Celtic *eburos with the non-Celtic words for ‘boar’ carries
much conviction. A proto-form *epros will regularly yield Celtic *ebros but not *eburos, so (1) fails.
The argumentation underlying (2) can only be characterized as obscurum per obscurius: this opaque
and isolated sound correspondence of Celtic voiced to non-Celtic voiceless plosives hardly has the
explanatory value required to convince us that *eburos is a more or less regular Celtic reflex of non-
Celtic *epuros. Moreover, the entire etymon is beset with difficulties because of the impossibility
12 Jullian 1920-19266, II 35-36.
13 Vendryes 1923, 476-477.
14 I have modernized Vendryes’ reconstructions.
15 A similarly conditional formulation can be found in Vendryes (1948, 281-282): “Il y avait un dieu Sanglier, dont
on possède plusieurs représentations figurées, et dont le nom, s’il ne se cache pas parfois sous une forme eburo- (issue
de *epro- et confondue alors avec le nom de l’if, R[evue] C[eltique], XL, 477), était Moccus.’ Note that Vendryes gives a
slightly different account of *eburos ‘boar’ from that in 1923.
mélanges en l’honneur de pierre-yves lambert 69
of reconstructing it as anything resembling Indo-European. To begin with, the a- or Latin aper is
problematic, and its only feasible explanation is on the basis of a Proto-Indo-European zero grade
*h1pros.16 That means that the Slavic and Germanic forms must be so-called vṛddhi-derivatives
(with full grade and o-stem suffix and meaning something like ‘belonging to a boar, boar-like
animal’) of *h1pro-. The v- in the Slavic form is unexplained: perhaps it was taken over from Proto-
Indo-European *wers- ‘male animal’, which survives in words like Latin verrēs ‘boar’? Nor is it clear
why Slavic has i-stem rather than o-stem inflection. Finally, the medial -u- of Germanic *eburaz
lacks an explanation: if anything it seems like a Germanic innovation, comparable to -u- in Gothic
miluks ‘milk’ and in Germanic *alus- ~ *alis- ‘ash tree’, rather than a common Celto-Germanic
innovation.
Given the total absence of evidence for the existence of Celtic *eburos ‘boar’ apart from the
Bordeaux altar, and given the formidable problems that would confront a connection of Celtic
*eburos ‘boar’, if that existed, with its non-Celtic counterparts, it is prudent to stick to what we do
know, viz that Celtic *eburos was a phytonym, and to apply Occam’s razor in order to establish that
it was only a phytonym.
That leaves us with the Bordeaux altar. Since it dates from the third century, when free traffic
of speakers not only of Latin and Celtic but also of Germanic (and many other languages) across
much of the Empire would have been possible for many generations, it is conceivable that the
homonymy of the Celtic phytonym *eburos and the Germanic word for ‘wild boar’ would have fed
the imagination of people to such an extent that they could easily associate Celtic names containing
Eburo- with the meaning ‘boar’ and reinterpret those names accordingly, in the same way that later
on Celtic Eburācum was reinterpreted as Old English Eofor-wīc ’boar-town’ and subsequently as
Old Norse *Jófur-wīk (> York).
4. Irish
At this point in the argument, it is still very much an open question whether the phytonym *eburos
originally meant ‘yew’ or not. The next step is to subject the etymon to closer scrutiny, in the hope
of tracing back the semantic history in greater detail. Let us begin with Irish.
OIr. ibar (m, o) means ‘yew’ and compellingly goes back to a proto-form *eburos. DIL lists a
number of instances in which ibar is qualified by a genitive and denotes different plants:
• iubhar beinne or craige, lit. ‘mountain ibar’, is the evergreen conifer juniper (Juniperus
communis)
• iubhar thalmhain, lit. ‘earth ibar’ denotes a kind of spleenwort (Asplenium), an evergreen
fern
• iubhar sléibhe, Modern Irish iúr sléibhe, literally means ‘mountain ibar’ and denotes the
labiate perennial wood sage (Teucrium scorodonia); this plant is evergreen unless cut back
by sharp frosts
• It is not clear exactly which plant is denoted by iubar lena lit. ‘meadow ibar’ but the
suggestion provided by ‘yew’, ‘juniper’, ‘spleenwort’ and ‘sage vel sim.’ is that it likewise
16 Schrijver 1991, 29, 31.
70 mélanges en l’honneur de pierre-yves lambert
denoted a plant that usually remained green during winter. Its Latin equivalents (Ballsamita,
Gamandrie in Archiv für celtische Lexikographie i 335 § 74. Pursibatum, Timus, ibid i 334
§ 30) may point to a labiate plant.
Iubhar lena occurs embedded in the following series: .i. bilur Muiri .i. fraech .i. ibur lena .i. fochluc
‘Mary’s watercress, heather, field ibar, fochluc (an unknown kind of edible aquatic plant like watercress)’.
17 Again the suggestion is of a series of rather different plants whose sole resemblance to one
another is that they appear to be winter-green plants. This is ethnobotanically interesting because
it indicates that to the medieval Irish the most striking property of ibar was its evergreen nature,
which allowed them to label other plants with that same property as ibar as well. At the same time,
the Irish motif of winter-greenness presents a problem to the etymologist, for the cognates of ibar
in other Celtic languages and elsewhere in Indo-European do not denote evergreen plants, as we
shall see in a moment.
5. Welsh
Middle and Modern Welsh efwr, North Welsh ewr (masculine and collective) denotes ‘hogweed,
cow parsnip’ (Heracleum sphondylium). Middle Welsh examples are almost completely limited to
the genre of nature poetry (references are to manuscripts, which are abbreviated according to the
conventions of GPC ).
C.30.44 cev ewur ‘cow-parsnip is hollow’ (description of winter)18
T.21.5 pan yw keu efwr ‘why cow-parsnip is hollow’19
R.1032.30 melyn eithin crin euwr ‘the gorse is yellow, the cow-parsnip is withered’20
R.1035.35-6 chwec euwr chwerthinat tonn ‘sweet is the cow-parsnip, laughing is the wave’ 21
The poem from the Llywarch Hen cycle about the names of his fallen sons contains a powerful
image, which evokes the similarity of six feet high, withered cow-parsley stems standing in a
meadow along a river to the skeletal remains of a group of perished warriors (C.40.11-13):
Handid haus im achuisson / oe adav ar lan awon / y gid a llv ewur lluydon
‘Complaints are easier for me / because of his being left by the bank of the
river / with a host of withered cow-parsley’22
While efwr is the exact formal counterpart of Irish ibar, the semantic difference between the two
is formidable. Hogweed is a perennial umbellifer that sends out long hollow flower stalks, which
form a conspicuous aspect of fertile moist meadows both in summer and in winter.
17 Archiv für celtische Lexikographie i. 334 § 30.
18 Jackson 1935b I.15.2; Jackson 1935a B.14.2; Jacobs 2012.
19 Haycock 2007, 117 line 139 + note.
20 Jackson 1935b II.9.2; Jackson 1935a, D.9.2; Jacobs 2012.
21 Jackson 1935b, VI.10.2; Jackson 1935a, H.10.2; Jacobs 2012.
22 Williams 1935, VIII.4; Rowland 1990, 413 (text), 473 (translation, here reproduced).
mélanges en l’honneur de pierre-yves lambert 71
There is an unexpected twist to the lack of correspondence between Welsh and Irish. The
Modern Irish name for hogweed is feabhrán. It does not occur in medieval Irish and its origin
is unknown. The diminutive ending -án is common in Irish plant names. If we assume that the
initial f- is secondary, as so often in Irish (e.g. Modern Irish fáinne ‘ring’ from Old Irish áinne), we
are left with a basic noun *eabhar that is strikingly similar to W efwr. Yet it cannot reflect *eburos
according to regular Irish sound changes. This suggests that *eabhar is a medieval borrowing from
Welsh efwr.
6. Breton
Middle Breton heuor, Modern Breton evor, ivor (collective) means ‘(alder) buckthorns’ (Frangula
alnus). The Vannetais dialect has a form ivo ‘id.’, which for unknown reasons lacks the final -r.
Middle and Early Modern Breton sources are lexical:23 Middle Breton Nomenclator p. 104:
heuor Lat. alnus, heuor du Lat. alnus nigra. This indicates a meaning ‘alder’ rather than ‘buckthorn’,
which given the similarity of their leaves is not surprising (this feature is also expressed in the
Modern English name, alder buckthorn). The Early Modern dictionary of Le P elletier (1752) lists
evor, envor, efor, evo. The Vannetais dictionary of L’Armerye (1744) on p. 35 has ivo ‘buckthorn’.
Modern Breton shows the varieties /evor/, /ivor/, /ivu/ and, with secondary gw-, /gwewr/.24 The
development e > i and the loss of final -r in some of these forms is unexplained. Middle Breton h- is
not pronounced and non-etymological (as in e.g. ha ‘and’).
Buckthorn is a large deciduous shrub that produces bunches of small greenish flowers, which
develop into berries that turn red and finally black. Buckthorn shows no obvious resemblance to
yew or hogweed, either in appearance or in its uses.
7. Continental Celtic
As already noted, a Continental Celtic element *Ebur(o)- is frequent in onomastics. Because they
have little to add to the discussion about the etymology beyond what has already been explained in
sections 2 and 3, I shall limit myself to giving a few examples.
Personal names: from Spain, especially in the area of the Vettones: Ebureinius (El Raso, Candeleda,
Ávila), Eburianus (Duratón, Segovia), Gsg. Eburi (Villamesías and Puerta de Santa Cruz, Cáceres);
perhaps in the unclear Celtiberian compound Ebursunos.25 Gaul: Εβουρος,26 Eburius (CIL XII
4154 [Calvisson], 617 [Bordeaux]), E[b]uria (CIL XII 4733 [Narbonne]), Eburila (CIL XIII 1228
[Bourges])
Tribal names: Eburones (approximately present-day Brabant and Limburg, Netherlands and
Belgium), Aulerci Eburovices (central France), as discussed.
Place names: Eburo-briga (Avrolles, Yolles, France), Eburo-dunum (Brünn, Germany; Yverdon,
Switzerland; Embrun, Hautes-Alpes, France; perhaps also Brno, Czech Republic), Eburo-magus
(Bram, Aude, France), Eburo-brittium (S-W Portugal). The early British Celtic names Eburācum
23 GIB 789-790.
24 Favereau 1993 s.v. evor.
25 Wodtko 2000, 103.
26 RIG II.1 G-88, Saint-Rémy-de-Provence.
72 mélanges en l’honneur de pierre-yves lambert
(York, UK; consistent -o- in Roman inscriptions presupposes a spoken Latin form, not surprisingly
for a Roman colonia and legionary fortress) and Eburo Caslum (Ravennas, to be read as *Castellum,
exact location unknown, Southern Scotland or Northumberland) belong to this group.27
Theonym: a Roman altar dedicated to a divinity Eburus recently surfaced in Spain (Cuevas de
Soria).28
As stated earlier, none of these names offer the semantic information required to pin down the
exact meaning of the phytonym *eburos. Ironically, the only information about what it meant
in Continental Celtic does not derive from Continental Celtic itself but from Germanic: the
element eber- in Modern High German Eber-esche ‘rowan, mountain ash, Sorbus aucuparia’ is
isolated in Germanic and was probably borrowed from Continental Celtic. Its earliest attestation
is 15th century: eber-boum.
8. Proto-Celtic. The formal reconstruction as Proto-Celtic *eburos does not present any problems,
so the discussion centres around the meaning of this form. As we observed earlier, the semantic
difference between Irish, Welsh and Breton is considerable. With the available evidence, the
following reconstruction can be attempted. What the tall perennial ’hogweed’ and the shrub
’buckthorn’ have in common is the fact that they have clusters of small greenish to white flowers.
The rowan, indirectly attested for Continental Celtic in German Eber-esche, eber-boum, has umbellike
heads of small white flowers (superficially like the true umbels of hogweed), which are followed
by red berries, while the buckthorn has red, later black berries. Since the only features that link
buckthorn (Breton) and rowan (Continental Celtic) on the one hand, and yew (Irish) on the other,
are that they are large plants carrying berries, that feature combination probably goes back to the
semantics of Proto-Celtic *eburos. The feature of having umbel-like heads of small whitish flowers
is present in hogweed (Welsh) and rowan (Continental Celtic), and with a little bit of fantasy also
in buckthorn (Breton), but definitely not in yew (Irish). There are no phenotypical characteristics
that link hogweed (Welsh) to yew (Irish). Hence the weight of the evidence is rather against Proto-
Celtic *eburo- meaning ‘yew’ or ‘hogweed’ because on that basis it is difficult to explain the others as
derived meanings. The evidence rather is in favour of it denoting a shrubby plant with umbel-like
clusters of whitish flowers which are followed by berries. On the basis of the latter, the meanings
‘hogweed’ (large plant with white flowers in an umbel), ‘buckthorn’ (woody plant with berries) and
‘rowan’ (woody plant with white flowers in an umbel and berries) can all be derived. The meaning
‘yew’ is the real outlier. In the absence of other phenotypical and technological similarities, ‘yew’
can be linked to the other meanings on the basis of its red berries (technically: arils), whose flesh is
edible, in contradistinction to the highly toxic seeds. In respect to fruit colour, edibility and treelike
habitus, ‘yew’ is closer to ‘rowan’ than to ‘buckthorn’.
All this leads to the following scenario. Proto-Celtic *eburos probably denoted ‘rowan, Sorbus
aucuparia’. This may be preserved in Continental Celtic on the evidence of the Germanic loan
eber-boum, Eber-esche. In Insular Celtic, a new word for ‘rowan’ was adopted: Old Irish cáerthann,
27 Sims-Williams 2006, 78; Rivet - Smith 1979, 357.
28 See the edition by Sanz et al 2011, who speak out in favour of a meaning ‘wild boar’ on pages 451-452. I am
indebted to Professor David Stifter for drawing my attention to this inscription.
mélanges en l’honneur de pierre-yves lambert 73
Middle Welsh kerddin, Modern Breton kerzhin.29 This set *eburos free to be used for other plants:
‘buckthorn’ in Breton because it too is woody and has berries; ‘hogweed’ in Welsh because of the
similar inflorescence; and ‘yew’ in Irish because it has red edible ‘berries’.
The use of ‘rowan’ in place names is as self-explanatory as the use of any other tree name in that
function. Its use in personal names can be compared to the Ogam Irish name MAQI CAIRATINI ,
MAQI CARATTINN ‘Son of Rowan’. In traditional popular beliefs of northern and central
Europe, rowan was regarded as protective against lightning and evil forces.30
9. Latin
Proto-Celtic *eburos ‘rowan (?)’ does not have a persuasive etymology outside Celtic. Pokorny
suggested that it reflects an earlier *erburo-, which he compares with an Indo-European root *erbhmeaning
‘brown’.31 The connection is completely arbitrary.
To my knowledge, it has not yet been observed that semantically and formally Celtic *eburolinks
up much better with Latin ebulus, ebulum, which means ‘dwarf elder, danewort’ (attested
since Cato). Danewort (Sambucus ebulus) is a close cognate of the shrubby elder (Sambucus nigra),
from which it differs by being a herbaceous perennial, which dies down every winter, and in that
it forms dense stands. Incidentally, those characteristics as well as the presence of divided, pinnate
leaves and its preferred habitat of fertile, moist soils link it particularly with hogweed (Welsh efwr).
Like most of the Celtic words, ebulus denotes a large plant with umbel-like heads of small whitish
flowers which are followed by red, then black, berries.
Formally, Lat. ebulus is usually reconstructed as *edh-lo-, however, and connected with the
synonymous Gaulish odocos attested in Marcellus of Bordeaux (Med. Lib. 7.13). This usually is
an unreliable source for Gaulish but in this case the existence of the word is confirmed by Latin
glosses odecus, odicus and by the West-Germanic loan from Gaulish, Old High German atuh,
Old Low German aduk ‘danewort’ < *oduk-. Spanish yezgo ‘danewort’ points to an alternative
form *eduko-, however, which occurs as educu in Latin glosses. This probably is a cross of Gaulish
*odokos with Latin ebulus, for which another example is the Latin gloss ebucone.32 The semantic
connection of *odokos with ebulus is semantically flawless but formally this is an intrinsically weak
root etymology.
Lat. ebulus and odocos are further connected with Balto-Slavic *edli- ‘spruce, fir’ in e.g. Czech
jedle, Russian el’, Lith. ẽglė, OPruss. addle (Walde-Hofmann 1938 I 389). The comparison is
semantically strained. More importantly, the Balto-Slavic forms are more attractively connected
with the root *ad- in Lith. adýti ‘to darn’, ãdata ‘needle’ (cf. Derksen 2008, 139).
André (1985, 193) follows Brüch in suggesting that Gaulish odocos may rather belong to the
root *od- ‘smell’ (Lat. odor etc.) than to Lat. ebulus on account of the unpleasant smell given off
by the leaves of danewort. Modern Irish has a probable cognate of this Gaulish *od- in odhrán, an
29 On these forms, see Schrijver 1997, 305-306.
30 Bächtold-Stäubli - Hoffmann - Krayer 1927-1942 s.v. Eberesche. See further MacKillop 1998, 330-331 on
supernatural uses of rowan reported in medieval Irish literature.
31 Pokorny 1948-1949, 251. For the root, see Walde-Pokorny 1927-1932 I 146.
32 Walde-Hofmann 1938 I 388.
74 mélanges en l’honneur de pierre-yves lambert
alternative name for cow-parsnip: it is a diminutive in -án based on a lost noun *odhar, which itself
reflects *od-V-ro- and contains the vegetable suffix *-Vr(o)- (see 10).
These plausible alternatives to the traditional etymology of ebulus open up the possibility
of connecting ebulus with Celtic *eburo-. Both can be reconstructed as (Italo-Celtic)*ebhuro- or
*eburo-. Latin -l- instead of -r- may be due to the influence of other plant names, such as populus,
opulus, corulus. No further etymology is available (Albanian bërshẽn ‘stinking juniper’ allows too
many possible reconstructions33 to be a reliable cognate).
10. Word-formation
Celtic *eburo- is one of a number of phytonyms that contain a vegetable suffix *‑Vr-:
(1) Irish curar, Welsh cylor(enn), Breton keler(enn), coloren ‘pignut’ < *kururā
(2) Irish biror, Welsh berwr, beryren, Breton beler(enn), Gallo-Latin berula ‘watercress’ (Rorippa
sp.) < *beruro-; this is probably cognate with Latin ferula ‘plant of the family Apiaceae, with a long
stem’, and with Irish ber, bir ‘water’, topur, tipra ‘source, well’ < Proto-Indo-European *bher(u)- ‘to
gush, boil’ (?)
(3) Irish semmar ‘clover’ < *se/immVrā, cf. Old Norse smári < *smai-r- or *smē-r-; cf. also unrelated
Germanic *klaib-ra- beside *klaiwa- ‘clover, Klee’.34
(4) Modern Irish feabhrán ‘cow-parsley’, odhrán ‘cow-parsley’, as discussed earlier.
11. Breton borrowings
Latin ebulus was probably borrowed into Breton as *evl. This may be represented in Old Southwest
British hob-aebl, with unclear first element, in the Leiden Leechbook.35 *evl is attested in the Middle
Breton singulative /evlenn/: eulenn ‘cest vng arbre, Lat. coluis’ (Catholicon versions a and b); efflen
plural effl ‘poplar, Populus’. Modern Breton forms are elw, elo, effl ‘poplars, aspens’. The semantic
development in Breton was probably triggered by confusion with a native word for ‘aspen’, which
survives in the isolated form Middle Breton ezlenn (Catholicon, manuscript version) ‘aspen’ < Late
Proto-British *aiθn-enn (compare Welsh aethnenn ‘aspen’). A meaning ‘buckthorn’, presumably by
confusion with evor, is attested for some instances of elw, efl (Duros 1991, 97).
Modern Breton evor ‘hellebore’, Middle Breton an euor guen ‘white hellebore, veratrum’
(Nomenclator p. 94) probably has nothing to do with *eburo-. It may represent a garbled form of
Lat. (h)elleborus, whose first syllable was reanalysed and detached as the Romance definite article,
while the rest of the word regularly yielded Romance *evor. Yet unfortunately no such form seems
to be attested in Romance.
12. The Celtic word for ‘yew’
If *eburo- originally denoted the rowan, it is no longer necessary to posit two Proto-Celtic words for
yew, *eburo- and *iwo- (Welsh yw, Old Irish éo). The latter means ‘yew’ in all its Celtic representatives
and can therefore stand as the only Proto-Celtic word for ‘yew’.
33 Demiraj 1997, 99.
34 Probably neither of these forms is Indo-European origin: see Schrijver 1997.
35 Falileyev - Owen 2005, 45.
mélanges en l’honneur de pierre-yves lambert 75
It is probably significant as well as confounding that cognates of *iwo- in other branches of
Indo-European show a similar connection between yew and a tree of the genus Sorbus, with clusters
of white flowers followed by red berries. Celtic *iwo- agrees with Germanic * īwā (e.g. Old High
German īwa, Old English īw, Old Norse ȳr)36 in denoting the yew. But elsewhere the etymon
denotes a Sorbus:
• Lithuanian ievà, Latvian iẽva ‘rowan’ (Sorbus aucuparia; their Slavic counterparts, e.g.
Russian íva, Serbo-Croat ȉva, mean ‘willow’, possibly on account of the young branches of
Sorbus and willow being used for wattling) < *(h1)eiHw-eh2
37
• Greek ὄη, ὄα, oἴη ‘service tree’ (Sorbus domestica) < *(H)oiHw-eh2
Note also Latin ūva ‘bunch of grapes’ < *(H)oiHw-eh2 and Armenian aigi ‘grape’ < *(H)oiHw-ijeh2-,
which are semantically connected to the aforementioned words by referring to clusters of edible
fruits. Finally, there is Hittite eyan- (neuter), which denotes an evergreen tree and reflects *(h1)e(h1)
i-on-, without the *-w- that can be found in the other cognates.
It is possible to assume that the original meaning of this item, ‘yew (berry)’, was preserved at the
geographical extremes of the Indo-European world (Hittite and Celtic, Germanic), while central
languages innovated by applying the term to other woody plants producing edible berries, but this
can be no more than tentative.
In light of the fact that the transition from ‘yew’ to Sorbus (or the other way round) apparently
occurred twice in the history of the ancient Indo-European languages, there must have been an
obvious semantic link between the two. The only link that comes to mind is the edible fruits.
That leads us to surmise that eating the fleshy arils of yew fruit, in spite of the high toxicity of all
other parts of the tree including the seed itself, was probably a commonplace activity in prehistoric
Europe.
References
Jacques André, « Noms de plantes gaulois ou prétendus gaulois dans les textes grecs et latins », Études
Celtiques 22, 1985, 179-198.
Hanns Bächtold-Stäubli, Eduard Hoffmann-Krayer (eds.), Handwörterbuch des deutschen Aberglaubens,
1927-1942, Berlin - Leipzig, de Gruyter.
Paul Courteault, « Un autel votif à la Tutelle, découverte à Bordeaux en 1921 », Revue des études classiques
24, 1922, 236-246.
Xavier Delamarre, Dictionnaire de la langue gauloise, 2003, Paris, Éditions Errance.
Bardhyl Demiraj, Albanische Etymologien, 1997, Amsterdam - Atlanta, Rodopi.
Rick Derks en, Etymological Dictionary of the Slavic Inherited Lexicon, 2008, Leiden - Boston, Brill.
DIL: Dictionary of the Irish Language, Compact edition, 1983, Dublin, Royal Irish Academy.
Fanch Duros, Herbarum vernaculi. Lexique du nom des plantes en breton, 1991, Quimperlé, Éditions La
Digitale.
36 Germanic also has a number of unexplained alternative forms, which seem to reflect * ībā (German Eibe, Dutch
ijf ), * īgwā (Old High German īga, īgo), * īχwā (Old Saxon īch, Old English ēoh). The short *i of the Celtic form can be
explained as shortening caused by Dybo’s rule (Schrijver 1991, 357).
37 Derksen 2008, 216.
76 mélanges en l’honneur de pierre-yves lambert
D. Ellis Evans, Gaulish Personal Names. A Study of Some Continental Celtic Formations, 1967, Oxford, Oxford
University Press.
Alexander Falileyev, Morfydd Owen, The Leiden Leechbook, 2005, Innsbruck, Innsbrucker Beiträge zur
Kulturwissenschaft.
Francis Favereau, Geriadur ar Brezhoneg a-vremañ, 1993, Morlaix, Skol Vreizh.
Emil Gamillscheg, Etymologisches Wörterbuch der französischen Sprache, 1969, Heidelberg, Winter.
GIB: Roparzh Hemon, Geriadur istorel ar Brezhoneg, 1958-1979, Quimper, Preder.
GPC: Geiriadur Prifysgol Cymru, 1950-2002, Cardiff, University of Wales Press.
Marged Haycock, Legendary Poems from the Book of Taliesin, 2007, Aberystwyth, Cambrian Medieval Celtic
Studies.
Kenneth H. Jacks on, Studies in Early Celtic Nature Poetry, 1935a, Cambridge, Cambridge University Press.
——, Early Welsh Gnomic Poetry, 1935b, Cardiff, University of Wales Press.
Nicholas Jacobs , Early Welsh Gnomic and Nature Poetry, 2012, Cardiff, Modern Humanities Research
Association.
Camille Jullian, Histoire de la Gaule, 6th edition, 1920-1926, Paris, Hachette.
James MacKillop, Dictionary of Celtic Mythology, 1998, Oxford, Oxford University Press.
Ranko Matas ović, Etymologial Dictionary of Proto-Celtic, 2009, Leiden - Boston, Brill.
Julius Pokorny, « Zur keltischen Namenkunde und Etymologie », Vox Romanica 10, 1948-1949, 220-267.
RIG II.1: Michel Lejeune, Recueil des inscriptions gauloises II.1: Textes gallo-étrusques. Textes gallo-latins sur
pierre, 1988, Paris, Éditions du CNRS.
RIG IV: Jean-Baptiste Colbert de Beaulieu, Brigitte Fischer, Recueil des inscriptions gauloises IV: Les légendes
monétaires, 1998, Paris, Éditions du CNRS.
A.L.F. Rivet, Colin Smith, The Place-Names of Roman Britain, 1979, London, Batsford.
Jenny Rowland, Early Welsh aga Poetry, 1990, Cambridge, D.S. Brewer.
Alberto Sanz Arag onés, Carlos Tab ernero Galán, Juan Pedro Benito Bataneo, Patrizia de Bernardo
Stempel, « Nueva divinidad céltica en un ara de Cuevos de Soria », Madrider Mitteilungen 52, 2011,
440-456.
Peter Schrijver, The Reflexes of the Proto-Indo-European Laryngeals in Latin, 1991, Amsterdam - Atlanta,
Rodopi.
——, « Some western European substratum words », in A. Lubotsky (ed.), Sound Law and Analogy, 1997,
Amsterdam-Atlanta, Rodopi, 305-6.
Patrick Sims-Williams, Ancient Celtic Place-Names in Europe and Asia Minor, 2006, Oxford - Boston,
Philological Society.
Joseph Vendryes, « Périodiques », Revue celtique 40, 1923, 473-499.
——, « La religion des celtes », inMana. Introduction à l’histoire des religions 2, volume III: Les religions de
l’Europe ancienne, 1948, Paris, Presses Universitaires de France, 237-320.
Alois Walde, J.B. Hofmann, Lateinisches etymologisches Wörterbuch, 1938, Heidelberg, Winter.
Alois Walde, Julius Pokorny, Vergleichendes Wörterbuch der indogermanischen Sprachen I-III, 1927-1932,
Berlin - Leipzig, de Gruyter.
Ifor Williams, Canu Llywarch Hen, 1935, Cardiff, University of Wales Press.
Dagmar Wodtko, Wörterbuch der keltiberischen Inschriften, Monumenta Linguarum Hispanicarum V.1,
2000, Wiesbaden, Reichert.