Version classiqueVersion mobile

Dérivation nominale et innovations dans les langues indo‑européennes anciennes

 | 
Alain Blanc
, 
Isabelle Boehm

Première partie. La dérivation depuis l'indo‑européen jusqu'aux langues attestées

Old Norse ‑yn (Proto‑Germanic *‑unjō) and the re‑analysis and spread of derivational morphology through semantic association

On Old Norse Fjǫrgyn ‘Earth(‑goddess)’ and Hlóðyn ‘id.’, Celtic Hercynia (silua) ‘Hercynian forest’, Vedic pŕ̥śni‑ ‘mother of the Maruts’, and Proto‑Indo‑European *perḱ‑ ‘colourful, spotted, dark’

Riccardo Ginevra

Résumé

Cet article traite de deux explications morphologiques et sémantiques nouvelles pour deux noms de la déesse de la terre en vieil islandais, Fjǫrgyn et Hlóðyn. Il montre avec de nouveaux arguments que le suffixe proto‑germanique *‑u‑njō‑ (vieil islandais ‑yn) a été constitué par réanalyse à partir d'un suffixe qui apparaissait dans les formations proto‑indo‑européennes du type de devī́‑ et que sa diffusion à l’intérieur du germanique a dû être liée à des développements analogiques apparus à l’occasion d’associations sémantiques. Ainsi, on rapporte Fjǫrgyn à un dérivé hérité reposant sur la racine *perḱ‑ « être coloré, être tacheté, être sombre », correspondant à la collocation largement répandue [sombreterre], avec des parallèles en germanique, en indo‑iranien et possiblement en celtique, tandis que Hlóðyn est expliqué comme l’aboutissement d’une formation analogique d’origine plus récente qui reflète tout de même la conception très ancienne de la terre comme la « porteuse » de l’univers.

Texte intégral

1. Introduction

  • 1 For valuable criticism, discussion, and help with this paper, I am indebted to Alcorac Alonso Déni (...)
  • 2 On the so‑called “suffix apophony” (Suffixablaut) or “suffix exchange” (Suffixtausch) attested wit (...)
  • 3 On PGmc *‑u‑/‑i‑njō‑ in general, cf. Casaretto 2004, pp. 330‑331; Meid 1967, pp. 119‑123. On the o (...)

1The rare Old Norse (ON) suffix ‑yn is exclusively attested by three theonyms, namely Fjǫrgyn, Hlóðyn, and Sígyn, and is currently analyzed as the outcome of Proto‑Germanic (PGmc) *‑unjō‑ and (transposed as) Proto‑Indo‑European (PIE) *‑n̥i̯éh2.1 Its variant with weak inflection ‑ynja (*‑unjō‑n‑) is equally unproductive and exclusively attested by three (probably inherited) formations, namely vargynja ‘she‑wolf’ (cf. Old English [OE] °wyrgen ‘id.’), apynja ‘she‑ape’ (cf. Old High German [OHG] affin ‘id.’), and ásynja ‘goddess’. Both variants ‑yn and ‑ynja must be traced back to PGmc *‑u‑njō‑, a suffix which, together with its “apophonic”2 variant *‑i‑njō‑, was productive for the formation of feminine, pertinentive, and collective denominatives, cf. inter alia Gothic (Goth.) Saur‑ini ‘Syrian woman’ (: Saur ‘Syrian’) and OHG mist‑un ‘dung heap’ (: mist ‘dung’). The Germanic suffix is in turn currently traced back to PIE *´‑n‑ih2/‑n̥‑i̯éh2, a complex suffix originally occurring in devī́‑ feminines (type Vedic [Ved.] rā́j‑ñ‑ī ‘queen’ beside masc. rā́j‑an‑ ‘king’). PIE *´‑n‑ih2/‑n̥‑i̯éh2 was re‑analyzed (already within PIE) as the additive feminine suffix *´‑nih2, attested e.g. by Greek (Gk) πότ‑νια ‘mistress’ and Ved. pát‑nī‑ ‘id.’, the reflexes of *pót‑nih2 ‘id.’ (: *pót‑i‑ ‘master’).3

2The assumed (ultimate) devī́‑ origin of PGmc *‑unjō‑, however, currently finds only partial support in the history of the three archaic formations which attest ON ‑yn.

  • 4 Goth. fairƕus is here analyzed as a secondary ‑u‑stem, as per Schaffner 2001, pp. 190‑194; aliter (...)
  • 5 Possible Germanic reflexes of this formation may rather be identified in ON fjǫrr ‘tree’ and OHG f (...)
  • 6 The ON name has been traced back to PGmc *fergunja‑ and identified as the Germanic counterpart to (...)

3ON Fjǫrgyn, also attested as a poetic term for ‘land’ (Oddrúnargrátr 116), is one of the names of the mother of the thunder‑god Thor (Vǫluspá 5610), a goddess who is elsewhere identified as Jǫrð ‘Earth’ (cf. Skáldskaparmál 24). Fjǫrgyn is the outcome of PGmc *fergunjō‑, which also underlies the medieval German toponym Fergunna (present‑day Ore Mountains between Germany and the Czech Republic), whereas similar formations Goth. fairguni “ὄρος” ‘mountain’ and OE firgen° ‘id.’ reflect a neuter *fergunj‑a‑ which was probably backformed from *fergunj‑ō‑ (Casaretto 2004, p. 331). The long noted correspondence of Fjǫrgyn with the (Latinized) Celtic toponym Hercynia (silua) ‘Hercynian forest’ (first attested in Caesar), explained by Elmar Seebold (1967, p. 118) as a Germanic‑Celtic isogloss due to language contact, has prompted the analysis of both as reflexes of *perku‑ni̯éh2‘oak‑forest’, a derivative of *perku‑ ‘oak’ (: Latin [Lat.] quercus); this interpretation, however, is problematic on several grounds (cf. Schaffner 2001, pp. 193‑194). Fjǫrgyn has also been connected with the reflexes of PGmc *ferhw‑a‑ ‘world, life, person’ (Goth. fairƕus “κόσμος”,4 ON fjǫr ‘life’, OE feorh ‘id., living person’, inter alia), which the same communis opinio has correspondingly traced back to a reflex of *perku‑ ‘oak’ as well,5 according to the cosmological conception of the “World‑Tree” (Meid 1984, pp. 98‑101); the semantics of Fjǫrgyn ‘Earth, land’ and *ferhw‑a‑ ‘world, life, person’, however, share a much closer connection with the concept [earth] than with [tree]. The detail of the alleged connection between Fjǫrgyn and the ON theonym Fjǫrgynn (Lokasenna 262; Skáldskaparmál 19) or Fjǫrgvinn (Gylfaginning 9), the name of the father (or husband, cf. Faulkes 2005, p. 166) of the goddess Frigg, must remain open at the current state of the research.6

4ON Hlóðyn occurs as a name of Thor’s mother as well (Vǫluspá 562), prompting a semantic interpretation as ‘Earth, land’, i.e. as a synonym of Fjǫrgyn. A word of unclear etymology, it has been connected with OE hlōð ‘booty, troop’ and Middle High German (MHG) luot ‘load, crowd’ (de Vries 1962, s.v.).

5ON Sígyn, the name of the wife of the god Loki (Vǫluspá 355), may be traced back to PGmc *sīg‑unjō‑ and PIE *sei̯k‑n̥‑i̯éh2, the generalized weak stem of *sei̯k‑én‑ih2‑/‑n̥‑i̯éh2‘she of the pouring’ (Ved. °sécanī‑, cf. upa‑sécanī‑ ‘pouring, pouring ladle’), a devī́‑ derivative of *sei̯k‑eno‑ ‘pouring’ (Ved. °sécana‑ ‘id.’), whose meaning exactly corresponds to Sigyn’s role within Norse mythology (Gylfaginning 50), namely the collecting and pouring out of a liquid (cf. Ginevra 2018).

6The theory of the devī́‑ origin of the PGmc suffix *‑u‑njō‑ thus currently finds support exclusively in the derivational history of the name Sígyn. In the present contribution, I will attempt to provide further evidence in support of this theory, while also identifying a proportional analogy triggered by semantic association as one of the causes for the spread of *‑u‑njō‑ within Germanic. Firstly, two alternative options for a new formal analysis of ON Fjǫrgyn and PGmc *fergunjō‑ as reflex of PIE *perḱ‑ ‘to be colourful, spotted, dark’, both supported by data from IE poetic phraseology and mythology, will be advanced (see 2), namely as the outcome of *perḱu‑nih2/‑ni̯éh2 ‘she of the Earth’ (see 2.1; this analysis finds support in Germanic and Iranian) and as the outcome of *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Colourful, Spotted, Dark One’ (see 2.2; this interpretation is supported by parallels in Old Norse, Vedic, and, possibly, Celtic). Finally, a new analysis of ON Hlóðyn will be proposed, namely as a relatively recent derivative of PGmc *hlōþ‑ō‑ ‘load’ reflecting the IE conception of the Earth as “bearer”; the possible path which may have led to its formation by analogy with a re‑analyzed reflex of PGmc *fergunjō‑ will also be discussed (see 3).

2. ON Fjǫrgyn ‘Earth, land’ (PGmc *fergunjō‑) and PIE *perḱ‑ ‘to be colourful, spotted, dark’

7In my opinion, ON Fjǫrgyn ‘Earth, land’ (PGmc *fergunjō‑) may be analyzed as a reflex of PIE *perḱ‑ ‘to be colourful, spotted, dark’ (on which cf. Widmer 2005, passim), a root attested, inter alia, by Ved. pŕ̥śni‑ ‘colourful, spotted’ and the Hesychian gloss πράκνον· μέλανα ‘black’ (both reflexes of *pr̥ḱ‑nó‑), Welsh erch ‘colourful, spotted, dark’ and Middle Irish erc ‘spotted (fish, cow)’ (both reflecting *perḱ‑ó‑), Lat. pulcer ‘beautiful’ (: *pr̥ḱ‑ró‑), and OHG faro ‘colourful, beautiful, outstanding’ (: *porḱ‑u̯ó‑). From the point of view of word‑formation, two alternative analyses are possible, namely (see 2.1) as a ‑nih2 derivative and (see 2.2) as an ‑ih2 derivative.

2.1. Fjǫrgyn as reflex of *perḱu‑ni̯éh2 ‘she of the Earth’ (PIE *pérḱ‑u‑ ‘Earth, land’)

8Let us start from the first possible analysis. ON Fjǫrgyn ‘Earth, land’ and PGmc *fergunjō‑ may be traced back to *perḱuni̯éh2, the generalized weak stem of *perḱu‑nih2/‑ni̯éh2 ‘she of the Earth’, a derivative of PIE *pérḱ‑u‑ ‘Earth, land’. The latter would reflect the substantivization of the adjective *pérḱ‑u‑/pr̥ḱ‑éu̯‑ ‘colourful, spotted, dark’, which may also underlie PGmc *ferhwa‑ ‘world, life, person’ and would perfectly match Proto‑Iranian *parću‑ ‘(home)land (of the Iranians)’ (both treated infra).

  • 7 This twofold semantics has a parallel in Classical Armenian cʿamakʿ ‘dry, dry land’, as pointed ou (...)
  • 8 Cf. the overview in West 2007, pp. 179‑180. The Hittite expression has been analyzed by Norbert Oe (...)

9The reconstructed PIE *pérḱ‑u‑ may be analyzed as a so‑called “transferred epithet” (Watkins 1995, p. 156) of the Earth, i.e. a designation which may be originally traced back to a traditional epithet, in the same way as Ved. Pr̥thivī́‑ ‘Earth’ literally means ‘the Broad One’ (being the feminine of the adjective pr̥thú‑ ‘broad’) and Lat. Terra ‘Earth, land’ is a reflex of *ters‑ā‑ ‘the Dry One’ (cf. *tērs‑os‑: Old Irish tír ‘earth, land’).7 A transferred epithet ‘the Dark One’ for the ‘Earth’ would match the phraseological collocation [dark – earth] attested in several Indo‑European languages, e.g. Hittite (KUB 17.10 ii 34 kat‑ta da‑an‑ku‑i te‑e‑kán za‑aḫ‑ḫi‑iš‑k[e‑e]z‑zi “below he strikes the dark earth”), Greek (Hes. Th. 69 […] περὶ δ’ ἴαχε γαῖα μέλαιναthe dark earth echoed”), Classical Armenian (Bowzandaran 3.14 yankarcōrēn jiwnn cʻamakʻ arǰn linēr aṙaǰi nora “suddenly, the snow in front of him became black earth”; Daniel Kölligan, pers. comm.), and Old Irish (Aigidecht Aithirne 1.10 ōs íath domuin duinn “over the territory of the brown world”).8

10As anticipated supra, this interpretation finds support in parallels in other Germanic and Indo‑European languages. Firstly, PGmc *ferhwa‑ ‘(earthly) world, life, (living) person’ may reflect *pérḱu̯‑o‑ ‘the Earthly One’, a substantivization of *perḱu̯‑ó‑ ‘earthly’, a possessive derivative of PIE *pérḱ‑u‑ ‘Earth, land’. This analysis may account for the various meanings of some of its reflexes:

    • 9 Adapted from King James Version; Greek original text: ἐξῆλθον παρὰ τοῦ πατρὸς καὶ ἐλήλυθα εἰς τὸν (...)

    Goth. fairƕus “κόσμος” in the Gospels always refers to our ‘world’, [earth], often in opposition to [heaven], cf. e.g. John 16:28 uzuhiddja fram attin jah atiddjain þana fairƕu; aftra bileiþaþamma fairƕaujah gagga du attin “I came forth from the Father (i.e. from Heaven), and am come into the world (i.e. Earth): again, I leave the world (i.e. Earth), and go to the Father (i.e. to Heaven)”.9

  • ON fjǫr ‘(earthly) life’ and OE feorh ‘id.’ may reflect the (fairly trivial) association of [living] with the act of [being/walking – on earth], widely attested in several IE languages (cf. the overview in West 2007, p. 125), e.g. in Old Norse (Dronke 1997, pp. 42‑43), cf. Oddrúnargrátr 81‑2 knátti mær oc mǫgr | moldveg sporna “a girl and a boy kicked the earth (i.e. they came to life)”.
  • OE feorh ‘living person, human being’ may reflect the conceptualization of [man] as an [earthly] sentient being (in opposition to a [deity], which is a [heavenly] sentient being, cf. Lat. deus ‘god’, Ved. devá‑ ‘id.’, reflexes of PIE *dei̯u̯‑ó‑ ‘heavenly’). This widespread association is attested, inter alia, by Germanic lexemes like OE guma and ON gumi ‘man’ (PGmc *guman‑: Goth. guma ‘id.’) and by Lat. homō ‘id.’, which all originally meant ‘the Earthly One’, being reflexes of an ‑n‑ derivative of PIE *dhéǵhom‑/dhǵhm̥‑´ ‘Earth, land’ (: Gk χθών ‘id.’).

11Secondly, the reconstructed formation *pérḱ‑u‑ ‘Earth, land’ underlying ON Fjǫrgyn (*perḱu‑nih2/‑ni̯éh2 ‘she of the Earth’) perfectly matches Proto‑Iranian *parću‑ ‘(home‑)land (of the Iranians)’, the substantivization of *pérḱ‑u‑/pr̥ḱ‑éu̯‑ ‘colourful, spotted, dark’ which has been reconstructed by Paul Widmer (2005, pp. 452‑453) as the ancestor of Pashto Pā̆ṣ̌, Puṣ̌t ‘(home‑)land of the Afghans’ and the derivational basis of various Iranian ethnic denominations, namely:

  • Old Persian Pārsa ‘Persian’ (cf. Assyr. Babyl. parsua, parsu/amaš ‘Persians’), the reflex of a vr̥ddhi formation *pārću̯a‑;
  • Old Persian Parθava‑ ‘Parthian’ and Inscriptional Pahlavi plswby /pahlav/ ‘id.’, which reflect *parćau̯(a);
  • Pashto pā̆ṣ̌to ‘Pashto’ and pā̆ṣ̌tūn ‘Pashtun’, the outcomes of *parć()‑āu̯a‑ and *parć()‑āna‑, respectively.
  • 10 I am also very grateful to Daniel Kölligan for pointing out to me the parallels between Widmer’s a (...)
  • 11 On these Germanic terms, cf. Lühr 2000, p. 60; Schaffner 2001, p. 191.

12As pointed out to me by Daniel Kölligan,10 these formations may have originally meant ‘the ones of the land’, i.e. ‘human beings, people’ (‘the Earthly Ones’ in opposition to the ‘heavenly’ deities); the employment of a term meaning ‘the people’ as an ethnonym is typologically frequent (Cardona 1989, pp. 350‑351). In this case, the Iranian derivatives would closely parallel the derivational history and semantics of North‑West‑Germanic (NWGmc) *firh(w)‑ija‑ ‘human beings, people’ (ON firar, fjǫrvar, fyrvar, OE fīras, OHG °firhi, gen.pl. fireo, Old Saxon gen.pl. firio),11 a derivative of PGmc *ferhwa‑ ‘world, life, person’ and thus another reflex of PIE *pérḱu‑ ‘Earth, land’.

13To sum up, ON Fjǫrgyn may be analyzed as a ‑nih2 derivative of PIE *pérḱ‑u‑ ‘Earth, land’, originally meaning ‘the Dark One’ and matching the phraseme [dark – earth]; this is supported by close parallels in Germanic (*ferhwa‑) and by an exact match in Iranian (*parću‑).

2.2. Fjǫrgyn as a reflex of *perḱn̥‑i̯éh2 ‘she, the Dark One’ (*pérḱn‑o‑ ‘the Dark One’)

  • 12 The adjective περκνός actually occurs in later sources as well; it may reflect a back‑formation af (...)
  • 13 On this derivational process, cf. Höfler 2017, pp. 131‑133 with literature.
  • 14 Cf. Middle Welsh elein ‘doe’, reflex of *h1él‑n‑ih2‑/‑n̥‑i̯éh2, derivative of *h1el‑nó‑ (Gk. ἑλλό (...)

14Alternatively, ON Fjǫrgyn (PGmc *fergunjō‑) may be analyzed as the expected outcome of PIE *perḱn̥‑i̯éh2, the weak stem of *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Dark One’, a devī́‑ feminine of PIE *pérḱn‑o‑ ‘the Dark One’ (cf. πέρκνος ‘dark eagle’, Aristarch’s emendation for Hom. περκνός in Il. 24.316).12 The latter would be a “vr̥ddhi substantivization”13 of PIE *pr̥ḱ‑nó‑ ‘colourful, spotted, dark’, the ‑nó‑ adjective of the root *perḱ‑ attested by Ved. pŕ̥śni‑ ‘colourful, spotted’ (on which cf. infra) and by the Hesychian gloss πράκνον· μέλανα ‘black’. A semantic shift from the expected meaning of *pérḱn‑ih2 ‘she, the Dark One’14 to the one attested for ON Fjǫrgyn ‘Earth, land’ would find support in the same phraseme [dark – earth] (see 2.1).

15This interpretation is supported by parallels in both Old Norse and other Indo‑European languages, namely Vedic and Celtic. Firstly, the ON poetic word fjǫrn ‘Earth’ (Skáldskaparmál 75 [501] jǫrð fjǫrn rufa […] “[poetic words for] earth: fjǫrn, rufa”) may reflect PGmc *ferhn‑ō‑, which may be transposed as *pérḱn‑eh2 ‘she, the Dark One’, a further feminine derivative of PIE *pérḱn‑o‑ ‘the Dark One’. The occurrence of Fjǫrgyn (‑ih2/‑i̯éh2 feminine) beside fjǫrn (‑eh2 feminine) parallels the attestation of ON Sígyn and Ved. °sécanī‑ ‘pouring, pouring‑ladle’ beside Celtic Sēquana ‘river‑goddess’ (the reflexes of *sei̯k‑én‑ih2‑/‑n̥‑i̯éh2and *sei̯ken‑eh2, respectively; Ginevra 2018) and that of Gk. ἄελλα ‘storm wind’ beside Middle Welsh awel ‘breeze’ (reflecting *h2eu̯h1el‑ih2 and *h2eu̯h1el‑eh2, respectively; Peters 1980, p. 195).

  • 15 On this goddess, cf. e.g. Macdonell 1897, p. 78.

16Secondly, the analysis of Fjǫrgyn as a reflex of PIE *pr̥ḱ‑nó‑ ‘colourful, spotted, dark’ finds support in Ved. Pŕ̥śni‑ ‘the Colourful, Spotted One’, the name of a goddess (RV +),15 which reflects the adjective pŕ̥śni‑ ‘colourful, spotted, speckled’ (RV +). The latter must be traced back to the re‑analysis of an abstract pŕ̥ś‑n‑i‑ ‘colourful‑ness, spotted‑ness’, an ‑i‑ derivative of the adjective *pr̥ś‑ná‑, the expected outcome of PIE *pr̥ḱ‑nó‑ ‘colourful, spotted, dark’; a similar process is currently assumed for Ved. bhū́ri‑ ‘much, many’ (cf. the same “substantival” accentuation as pŕ̥śni‑), which, as demonstrated by Georges‑Jean Pinault (1998), must be traced back to the re‑analysis of an abstract *bhū́‑r‑i‑ ‘abundance’, an ‑i‑ derivative of the adjective *bhū‑rá‑ ‘much, many’.

17The analysis of both ON Fjǫrgyn and Ved. Pŕ̥śni‑ as reflexes of PIE *pr̥ḱ‑nó‑ ‘colourful, spotted, dark’ finds support in their shared semantic associations, as attested by poetic phraseology:

(a) Fjǫrgyn and Pr̥śni are both referred to as [mother – of thunder‑god]

18ON Fjǫrgyn is a name of the Norse earth‑goddess, the mother of the thunder‑god Thor (Þórr: PGmc *þunara‑ ‘thunder’; cf. Harðarson 2001, pp. 104‑105):

Skáldskaparmál 24
Hvernig skal jǫrð kenna? Kalla […] móður Þórs […].

How shall earth be referred to? By calling it mother of Thor.

Vǫluspá 569‑10
gengr fet nío    Fiorgyniar burr

he steps nine paces, Fjǫrgyn [= Earth]’s child [i.e. Thor].

19Ved. Pŕ̥śni‑ is the name of the mother of the Maruts, the Vedic gods of the thunderstorms:

RV 1.23.10b‑11b
marútaḥ sómapītaye
ugrā́ hí pŕ̥śnimātaraḥ
jáyatām iva tanyatúr
marútām eti dhr̥ṣṇuyā́

[We call] the Maruts, for soma‑drinking, for they are strong, with Pr̥śni as their mother. The thundering of the Maruts, like that of victors, goes boldly.

(b) Fjǫrgyn and Pr̥śni are both identified with [earth]

20ON fjǫrgyn is a poetic synonym of jǫrð ‘earth’:

Oddrúnargrátr 116
enn ec fylgðac þér    á fiǫrgynio

And I used to accompany you on earth.

21Ved. Pŕ̥śni‑ appears to be employed as a synonym of pr̥thivī́‑ ‘earth’ in the Atharvavedic hymn To the plants, where they are both referred to as “mother” of the plants:

AVŚ 8.7.2cd
yā́sām dyáuḥ pitā́ pr̥thivī́ mātā́ samudró mū́laṃ vīrúdhāṃ babhū́va

The plants of which heaven has been the father, earth the mother, ocean the root.

AVŚ 8.7.21cd
yadā́ vaḥ pr̥śnimātaraḥ parjányo rétasā́vati

When Parjanya favors you [plants] who have Pr̥śni as their mother with seed.

  • 16 Cf. also Jamison 1991, p. 263, on the Vedic association of the Earth with a pŕ̥śni‑ ‘spotted’ cow, (...)

22Moreover, Pr̥śni seems to be identified with the sacrificial ground (or at least with parts of it) in some passages of the RV (Jamison, Brereton 2014, ad 8.94) and was equated with [earth] by the commentator Sāyaṇa (ad RV 2.34.2).16 The characterisation of Fjǫrgyn and Pr̥śni as both (a) mothers of thunder‑gods and (b) earth‑goddesses has parallels in the conceptualization of [earth] as the origin of [thunder] occurring in Greek mythology:

Hes. Th. 504‑505
δῶκαν δὲ βροντὴν ἠδ᾽ αἰθαλόεντα κεραυνὸν
καὶ στεροπήν· τὸ πρὶν δὲ πελώρη Γαῖα κεκεύθει

  • 17 Perhaps to be connected with the phenomenon of “seismic thunder” (West 1966, ad loc.), cf. A. Pr. (...)

They [the Cyclopes] gave him [to Zeus] the thunder and the blazing thunderbolt and the lightning, which huge Earth had concealed before.17

23The phraseological matches between ON Fjǫrgyn and Ved. Pŕ̥śni‑ may allow for the reconstruction of an inherited mythical character, which was characterized as ‘colourful, spotted, dark’ (*pr̥ḱ‑nó‑), regarded as a mother of thunder‑gods, and identified with the Earth (the origin of thunder).

24Finally, a third parallel to the analysis of Fjǫrgyn as the outcome of *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Dark One’ may be provided by the Celtic toponym Hercynia (Silua). Caesar explicitly tells us that his source for the toponym was not a Greek literary source:

Caes. Gall. 6.24
[…] circum Hercyniam silvam, quam Eratostheni et quibusdam Graecis fama notam esse video, quam illi Orcyniam appellant […].

[…] round the Hercynian forest (which I see was known by report to Eratosthenes and certain Greeks, who call it the Orcynian forest) […].

25It is thus conceivable that, with his notation Hercynia, Caesar was actually trying to approximate a native informant’s pronunciation /(h)erkəni̯ā‑/, which may reflect a Celtic dialectal outcome of PIE *perḱn̥‑i̯éh2, the weak stem of *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Dark One’. In this case, Celtic Hercynia ‘she, the Dark One’ would turn out to be, on the one hand, a perfect match for ON Fjǫrgyn and, on the other hand, a very close parallel for various medieval and modern German toponyms referring to locations which were once part of the Hercynian forest, namely:

  1. the already mentioned Ore Mountains (German Erzgebirge, Czech Krušné hory), a forested mountain range between Germany and the Czech Republic, referred to in medieval sources both as Fergunna (PGmc *fergunjō‑, another reflex of PIE *perḱn̥‑i̯éh2 ‘she, the Dark One’) and as Miriquidi, matching ON Myrk‑viðr ‘Dark Forest’, the name of various mythical and historical Scandinavian forests (Rübekeil 1992, pp. 69‑70; 2002, p. 103);
    • 18 I’m grateful to Marek Majer for pointing out the Polish toponyms to me.

    the Black Forest, a forested mountain range in South‑West Germany, whose Modern German name Schwarz‑wald ‘Black‑forest’ semantically matches the formations above.
    The fairly trivial toponomastic motif [dark – forest] is also attested e.g. by ItalianSelva Nera (near Rome), French Bois Noirs (in Livradois), CzechČerný les (name of various forests), as well as Polish Czarny Las and Czarnolas (names of several localities).18

26To sum up, ON Fjǫrgyn may alternatively reflect *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Dark One’, an ‑ih2 derivative of PIE *pérḱ‑no‑ ‘the Dark One’, with parallels in Old Norse (fjǫrn), Vedic (Pŕ̥śni‑) and, possibly, Celtic (Hercynia).

3. ON Hlóðyn as a reflex of NWGmc *hlōþ‑unju ‘the one of the load’ (PGmc *hlōþ‑ō‑ ‘load’)

  • 19 Cf. OE fōr ‘travel’: faran ‘to go’ (Bammesberger 1990, p. 111).
  • 20 On PGmc *hlaþ‑a‑ ‘to load’ and its derivatives, cf. Seebold 1970, s.v.
  • 21 Cf. the overview in West 2007, p. 179.

27Let us now turn our attention to ON Hlóðyn ‘Earth’ and to one possible itinerary for the re‑analysis and spread of the suffix *‑unjō‑ in Germanic. ON Hlóðyn may be analyzed as the expected outcome of PGmc *hlōþ‑unjō‑ ‘she of the load’, a pertinentive derivative of *hlōþ‑ō‑ ‘load’ (OE hlóð ‘booty, troop’, Old Frisian hlōth, MHG luot ‘burden, crowd’). The latter is a derivative (of the same type of PGmc *fōr‑ō‑ ‘travel’: *far‑a‑ ‘to go’)19 of the verb *hlaþ‑a‑ ‘to load’ (Goth. ptc. °hlaþans, ON hlaða, OE hladan).20 The semantic shift from PGmc *hlōþ‑unjō‑ ‘she of the load’ to ON Hlóðyn ‘Earth’ would find support in the IE cosmological conception of the Earth as a “bearer”,21 reflected by the phraseological collocation [earth(‑goddess) – bear – world, all beings] attested e.g. in Vedic, Avestan, and Greek:

RV 7.34.7b
bíbharti bhārám pr̥thivī́ bhū́ma

It bears its burden, like the earth [bears] its ground.

AVŚ 5.28.5
bhū́miṣ ṭvā pātu háritena viśvabhŕ̥d […].

Let earth, the allbearing, protect thee with the yellow one […].

Y. 38.1
imąm […] ząm […] nā̊ baraitī […].

This earth that bears us [all].

A. Pers. 618
ἄνθη τε πλεκτά, παμφόρου Γαίας τέκνα

  • 22 Literal meaning; in this passage, however, Gk. φέρω ‘bear’ is rather used “of the earth ‘bearing’ (...)

a woven garland of flowers, the children of Earth, the bearer of all.22

28Given the fact that the suffix *‑unjō‑ was not particularly productive for the formation of theonyms, however, the parallel between the two synonyms Fjǫrg‑yn and Hlóð‑yn can hardly be a coincidence.

29A possible solution may be to trace ON Hlóðyn back to a NWGmc formation *hlōþ‑unju built by semantic association with a reflex of PGmc *fergunjō‑. This process may have taken place in two steps: 

    • 23 For this analysis of ON fjǫrg, cf. Schaffner 2001, p. 193. NWGmc *ferg‑u ‘world, living beings’ re (...)

    Re‑analysis of NWGmc *fergunju ‘Earth’ (PGmc *fergunjō‑) as *ferg‑unju ‘she of the world’.
    NWGmc *fergunju ‘Earth’, the expected outcome of PGmc *fergunjō‑ and PIE *perḱu‑ni̯éh2 or *perḱn̥‑i̯éh2, could be easily re‑analyzed as ferg‑unju ‘she of the world’, a derivative of the neut.pl. *ferg‑u ‘world, living beings’ (ON fjǫrg ‘id.’ in Lokasenna 196). This was in turn the outcome of PGmc *fergw‑ō‑ ‘id.’, the Vernerized collective plural of PGmc *ferhw‑a‑ ‘world, life, person’.23

  1. Formation of NWGmc *hlōþ‑unju ‘Earth (she of the load)’ by semantic association with *ferg‑unju.
    In a second step, NWGmc *hlōþ‑unju ‘Earth (she of the load)’ was derived from *hlōþ‑u ‘load’ (PGmc *hlōþ‑ō‑ ‘id.’), by means of a proportional analogy triggered by semantic association with *ferg‑unju ‘Earth’ (which had been re‑analyzed as ‘she of the world’) and *ferg‑u ‘world, living beings’:
          NWGmc *ferg‑unju : *ferg‑u ∷ x : *hlōþ‑u, whence x = *hlōþ‑unju.

4. Conclusion

30The results of this study may be summarized as follows:

  1. The evidence presented confirms the PIE devī́‑ origin of the PGmc suffix *‑u‑njō‑ and points to semantic association as one of the causes for its spread within the lexicon.
  2. ON Fjǫrgyn ‘Earth, land’ (PGmc *fergunjō‑), may be traced back to PIE *perḱ‑ ‘to be colourful, spotted, dark’ either as a reflex of (a) *perḱu‑nih2/‑ni̯éh2 ‘she of the Earth’, a derivative of PIE *pérḱ‑u‑ ‘Earth, land’, originally ‘colourful, spotted, dark’, matching the IE phraseological collocation [dark – earth], with parallels in Germanic (*ferhwa‑ ‘world, life, person’) and Iranian (*parću‑ ‘[home]land [of the Iranians]’), or as a reflex of (b) *pérḱn‑ih2/perḱn̥‑i̯éh2 ‘she, the Colourful, Spotted, Dark One’ (reflecting the same collocation [dark – earth]), a feminine derivative of *pérḱn‑o‑ ‘the Dark One’ (a substantivization of *pr̥ḱ‑nó‑ ‘colourful, spotted, dark’), with parallels in Old Norse (fjǫrn ‘Earth, land’), Vedic (Pŕ̥śni‑ ‘the Colourful, Spotted One’) and, possibly, Celtic (Hercynia). Both options fit into the current view that PGmc *‑u‑njō‑ reflects PIE nih2ni̯éh2, either (a) already re‑analyzed as an independent suffix or (b) as a complex suffix occurring in devī́‑ feminines.
  3. ON Hlóðyn ‘Earth’ is the outcome of NWGmc *hlōþ‑unju ‘the one of the load’, which reflects the IE conception of the Earth as “bearer”, attested by the collocation [earth(‑goddess) – bear – world, all beings]. NWGmc *hlōþ‑unju was formed from *hlōþ‑u ‘load’ (PGmc *hlōþ‑ō‑ ‘id.’) after the model of NWGmc *ferg‑unju (ON Fjǫrgyn), the outcome of PGmc *fergunjō‑ which had been re‑analyzed as a derivative of NWGmc *ferg‑u ‘world, living beings’ (ON fjǫrg).

Bibliographie

Bammesberger 1990: A. Bammesberger, Die Morphologie des urgermanischen Nomens, Heidelberg, Winter, 1990.

Beekes 2010: R. Beekes, Etymological dictionary of Greek, Leiden/Boston, Brill, 2010.

Cardona 1989: G.R. Cardona, “Noms de peuples et noms de langues”, in G. Calame‑Graiule (ed.), Graines de paroles, Paris, Éditions du CNRS, 1989, pp. 349‑360.

Casaretto 2004: A. Casaretto, Nominale Wortbildung der gotischen Sprache. Die Derivation der Substantive, Heidelberg, Winter, 2004.

De Vries 1962: J. de Vries, Altnordisches etymologisches Wörterbuch, Leiden, Brill, 1962.

Dronke 1997: U. Dronke, The poetic Edda, vol. II, Mythological poems, Oxford, Clarendon Press, 1997.

Edwards 1917: H.J. Edwards, Caesar: the Gallic war, Cambridge, Harvard University Press, 1917.

Faulkes 1987: A. Faulkes, Edda. Snorri Sturluson. Translated from the Icelandic and introduced, London, Dent, 1987.

Faulkes 2005: A. Faulkes, Edda. Prologue and Gylfaginning, London, Viking Society for Northern Research, 2005 (2nd ed.).

Faulkes 2007: A. Faulkes, Edda. Skáldskaparmál, London, Viking Society for Northern Research, 2007 (2nd ed.).

Ginevra 2018: R. Ginevra, “Old Norse Sígyn (*sei̯k‑n̥‑i̯éh2 ‘she of the pouring’), Vedic °sécanī‑ ‘pouring’, Celtic Sēquana and PIE *sei̯k ‘pour’”, in D.M. Goldstein, S.W. Jamison, B. Vine (ed.), Proceedings of the 29th Annual UCLA Indo‑European conference, Bremen, Hempen Verlag, 2018, pp. 65‑76.

Harðarson 2001: J.A. Harðarson, Das Präteritum der schwachen Verba auf ‑ýia im Altisländischen (und verwandte Probleme der altnordischen und germanischen Sprachwissenschaft), Innsbruck, Institut für Sprachen und Literaturen der Universität Innsbruck, 2001.

Hoffner 1998: H.A. Hoffner, Hittite myths, Atlanta, Scholars Press, 1998 (2nd ed.).

Höfler 2017: S. Höfler, Der Stier, der Stärke hat. Possessive Adjektive und ihre Substantivierung im Indogermanischen, PhD, University of Vienna, 2017 (unpublished).

Jamison 1991: S.W. Jamison, The ravenous hyenas and the wounded sun: myth and ritual in ancient India, Ithaca/London, Cornell University Press, 1991.

Jamison, Brereton 2014: S.W. Jamison, J.P. Brereton, The Rigveda. The earliest religious poetry of India, Oxford/New York, Oxford University Press, 2014.

Kroonen 2013: G. Kroonen, Etymological dictionary of Proto‑Germanic, Leiden, Brill, 2013.

Kulišić, Petrović, Pantelić 1970: S. Kulišić, P.Ž. Petrović, N. Pantelić, Srpski Mitološki Rečnik, Beograd, Nolit, 1970.

Lühr 2000: R. Lühr, Die Gedichte des Skalden Egill, Dettelbach, Röll, 2000.

Macdonell 1897: A.A. Macdonell, Vedic mythology, Strasbourg, Trübner, 1897.

Meid 1967: W. Meid, Germanische Sprachwissenschaft, III. Wortbildungslehre, Berlin, De Gruyter, 1967.

Meid 1984: W. Meid, “Bemerkungen zum indogermanischen Wortschatz des Germanischen”, in J. Untermann, B. Brogyanyi (ed.), Das Germanische und die Rekonstruktion der indogermanischen Grundsprache. Akten des Freiburger Kolloquiums der Indogermanischen Gesellschaft, Freiburg, 26.‑27. Februar 1981, Amsterdam/Philadelphia, Benjamins, 1984, pp. 91‑112.

Most 2007: G.W. Most, Hesiod. Theogony. Works and days. Testimonia, Cambridge/London, Harvard University Press, 2007.

Neri 2003: S. Neri, I sostantivi in ‑u del gotico. Morfologia e preistoria, Innsbruck, Institut für Sprachen und Literaturen der Universität Innsbruck, 2003.

Oettinger 1989‑1990: N. Oettinger, “Die ‘dunkle Erde’ im Hethitischen und Griechischen: Alfred Heubeck zum Gedächtnis (20.7.1914‑24.5.1987)”, Die Welt des Orients 20/21, 1989‑1990, pp. 83‑98.

Peters 1980: M. Peters, Untersuchungen zur Vertretung der indogermanischen Laryngale im Griechischen, Vienna, Verlag der Österreichischen Akademie der Wissenschaften, 1980.

Pinault 1998: G.‑J. Pinault, “Védique bhū́ri‑, un ancien substantif”, Bulletin des Études Indiennes 16, 1998, pp. 89‑121.

Ringe, Taylor 2014: D. Ringe, A. Taylor, The development of Old English, Oxford, Oxford University Press, 2014.

Rübekeil 1992: L. Rübekeil, Suebica. Völkernamen und Ethnos, Innsbruck, Institut für Sprachen und Literaturen der Universität Innsbruck, 1992.

Rübekeil 2002: L. Rübekeil, Diachrone Studien zur Kontaktzone zwischen Kelten und Germanen, Vienna, Verlag der Österreichischen Akademie der Wissenschaften, 2002.

Schaffner 1996: S. Schaffner, “Zur Wortbildung und Etymologie von altenglisch nihol, niowol und lateinisch procul”, MSS 56, 1996, pp. 131‑171.

Schaffner 2001: S. Schaffner, Das Vernersche Gesetz und der innerparadigmatische grammatische Wechsel des urgermanischen im Nominalbereich, Innsbruck, Institut für Sprachen und Literaturen der Universität Innsbruck, 2001.

Seebold 1967: E. Seebold, “Die Vertretung von idg. guh im Germanischen”, Kuhns Zeitschrift 81, 1967, pp. 104‑133.

Seebold 1970: E. Seebold, Vergleichendes und etymologisches Wörterbuch der germanischen starken Verben, The Hague/Paris, Mouton, 1970.

Sommerstein 2008: Aeschylus, Persians, Seven against Thebes, Suppliants, Prometheus bound, ed. and trans. A.H. Sommerstein, Cambridge/London, Harvard University Press, 2008.

Watkins 1995: C. Watkins, How to kill a dragon: aspects of Indo‑European poetics, New York/Oxford, Oxford University Press, 1995.

Weiss 2017: M. Weiss, “King. Some observations on an East‑West archaism”, in B.S.S. Hansen, A. Hyllested, A.R. Jørgensen, G. Kroonen, J.H. Larsson, B. Nielsen Whitehead, T. Olander, T. Mosbæk Søborg (ed.), Usque Ad Radices, Indo‑European studies in honour of Birgit Anette Olsen, Copenhagen, Museum Tusculanum Press, 2017, pp. 415‑426.

West 1966: M.L. West, Hesiod, Theogony, Oxford, Oxford University Press, 1966.

West 2007: M.L. West, Indo‑European poetry and myth, Oxford, Oxford University Press, 2007.

Whitney 1905: W.D. Whitney, The Atharva‑Veda Saṃhitā: translated with a critical and exegetical commentary, revised and brought nearer to completion and edited by Charles Rockwell Lanman, Cambridge, Harvard University Press, 1905.

Widmer 2005: P. Widmer, “Etymologisches und Historisches zum Persernamen”, in V. Sadovski, D. Stifter (ed.), Iranistische und indogermanistische Beiträge in memoriam Jochem Schindler (1944‑1994), Vienna, Verlag der Österreichischen Akademie der Wissenschaften, 2005, pp. 445‑459.

Zair 2012: N. Zair, The reflexes of the Proto‑Indo‑European laryngeals in Celtic, Leiden, Brill, 2012.

Notes

1 For valuable criticism, discussion, and help with this paper, I am indebted to Alcorac Alonso Déniz, Cristina Alù, Lucien van Beek, Andrea Lorenzo Covini, Paola Dardano, Andrea Filoni, Romain Garnier, Matilde Garré, Daniel Kölligan, Marek Majer, Birgit Olsen, Rostislav Oreshko, Brent Vine, and especially José Luis García Ramón. Thanks are also due to Robert Tegethoff for improving my English version. The usual disclaimers apply.

The translations are adapted from Dronke 1997 (Vǫluspá); Edwards 1917 (Caesar); Faulkes 1987 (Skáldskaparmál); Hoffner 1998 (KUB 17.10); Jamison, Brereton 2014 (Rigveda); Most 2007 (Hesiod); Sommerstein 2008 (Aeschylus) and Whitney 1905 (Atharvaveda‑Śaunakīya).

Standard abbreviations are employed for Greek and Latin sources, as well as for Atharvaveda‑Śaunakīya (AVŚ) and Rigveda (RV). Passages from Snorri Sturluson’s Skáldskaparmál are quoted on the basis of Faulkes’ (2007) normalized edition.

2 On the so‑called “suffix apophony” (Suffixablaut) or “suffix exchange” (Suffixtausch) attested within Germanic derivational morphology, cf. Meid 1967, pp 50‑51; Schaffner 1996, pp. 150‑151.

3 On PGmc *‑u‑/‑i‑njō‑ in general, cf. Casaretto 2004, pp. 330‑331; Meid 1967, pp. 119‑123. On the origin of PIE *´‑n‑ih2/‑n̥‑i̯éh2, cf. Casaretto 2004, p. 330; Weiss 2017, p. 799.

4 Goth. fairƕus is here analyzed as a secondary ‑u‑stem, as per Schaffner 2001, pp. 190‑194; aliter Neri 2003, pp. 202‑207, and Casaretto 2004, pp. 194‑195; cf. also Lühr 2000, pp. 72‑73.

5 Possible Germanic reflexes of this formation may rather be identified in ON fjǫrr ‘tree’ and OHG fereh‑eih ‘id.’ (Kroonen 2013, s.v. *ferhwa‑ 1).

6 The ON name has been traced back to PGmc *fergunja‑ and identified as the Germanic counterpart to the Lithuanian theonym Perkū́nas and to various Indo‑European names of thunder‑gods which attest initial *per‑, *per‑k‑, or *per‑ǵ‑ (cf. the overview in West 2007, pp. 238‑245). The Norse character, however, is not associated with the concept [thunder] in any source (he is only mentioned in relation to the goddess Frigg, who shares no similar associations). Moreover, given that ON ‑yn‑ may reflect earlier ‑vin‑ (cf. the ON name of the city of Bergen in Norway, Bjǫrg‑yn, the reflex of earlier Bjǫrg‑vin ‘mountain‑meadow’), whereas the opposite development (‑vin‑ from ‑yn‑) is unparalleled, the original nom.sg. is likely to be Fjǫrgvinn (rather than Fjǫrgynn), which cannot reflect PGmc *fergunja‑.

7 This twofold semantics has a parallel in Classical Armenian cʿamakʿ ‘dry, dry land’, as pointed out to me by Daniel Kölligan (pers. comm.).

8 Cf. the overview in West 2007, pp. 179‑180. The Hittite expression has been analyzed by Norbert Oettinger (1989‑1990) as a calque from Hurrian which would have been later introduced into Greek as well. Given that this collocation is attested in several Indo‑European traditions (Anatolian, Greek, Armenian, Celtic, Slavic, and Baltic), however, there seems to be no reason to doubt that it may have already been Proto‑Indo‑European (Oettinger himself [1989‑1990, p. 92] admits that an origin as calque for the Greek formula “cannot be proved in strict sense” [“sich […] nicht im strengen Sinne beweisen läßt”]). On the contrary, these expressions seem to perfectly fit into a system of binary oppositions between [earth] and [heaven] (as the two ‘world‑halves’, Ved. rodasī́) which is widely attested in Indo‑European traditions (and probably universal), also underlying e.g. the conceptualization of [earth] as a [mother] in opposition to [father – heaven] (cf. the opposite conceptualization in Egyptian mythology: male [earth]‑god Geb vs female [heaven]‑goddess Nut) or the association of [earth] with [death/mortality] in opposition to that of [heaven] with [immortality]. Within this system, [earth] may have been referred to as [dark/brown] in opposition to [heaven], which is often conceptualized as [bright/white], cf. e.g. Serbian ne vidi beloga boga “(s)he does not see the white god” and Bulgarian vika do beloga boga “scream at the white god”, expressions in which the collocation [white – god] may stand for [heaven] (Kulišić, Petrović, Pantelić 1970, s.v. beli bog).

9 Adapted from King James Version; Greek original text: ἐξῆλθον παρὰ τοῦ πατρὸς καὶ ἐλήλυθα εἰς τὸν κόσμον· πάλιν ἀφίημι τὸν κόσμον καὶ πορεύομαι πρὸς τὸν πατέρα.

10 I am also very grateful to Daniel Kölligan for pointing out to me the parallels between Widmer’s analysis of the Iranian data and my own analysis of the Germanic data.

11 On these Germanic terms, cf. Lühr 2000, p. 60; Schaffner 2001, p. 191.

12 The adjective περκνός actually occurs in later sources as well; it may reflect a back‑formation after the pattern attested e.g. by λεῦκος ‘name of a (white) fish’: λευκός ‘white’ (substantive *CéC‑o‑: adj. *CeCó‑).

13 On this derivational process, cf. Höfler 2017, pp. 131‑133 with literature.

14 Cf. Middle Welsh elein ‘doe’, reflex of *h1él‑n‑ih2‑/‑n̥‑i̯éh2, derivative of *h1el‑nó‑ (Gk. ἑλλός ‘deer’; Zair 2012, pp. 195 and 248); cf. also Ved. devī́‑ itself, reflex of *déi̯u̯‑ih2‑ ‘goddess’ (‘she, the Heavenly One’), derivative of *dei̯u̯‑ó‑ ‘god’ (‘the Heavenly One’).

15 On this goddess, cf. e.g. Macdonell 1897, p. 78.

16 Cf. also Jamison 1991, p. 263, on the Vedic association of the Earth with a pŕ̥śni‑ ‘spotted’ cow, attested inter alia in Aitareya Brāhmaṇa 5.23.

17 Perhaps to be connected with the phenomenon of “seismic thunder” (West 1966, ad loc.), cf. A. Pr. 993 βροντήμασι χθονίοις.

18 I’m grateful to Marek Majer for pointing out the Polish toponyms to me.

19 Cf. OE fōr ‘travel’: faran ‘to go’ (Bammesberger 1990, p. 111).

20 On PGmc *hlaþ‑a‑ ‘to load’ and its derivatives, cf. Seebold 1970, s.v.

21 Cf. the overview in West 2007, p. 179.

22 Literal meaning; in this passage, however, Gk. φέρω ‘bear’ is rather used “of the earth ‘bearing’ its produce” (West 2007, p. 179), i.e. in its figurative meaning ‘bringing forth’; as suggested to me by Daniel Kölligan, the motive of the “Earth as bearer” is more clearly attested in the Hom. formula ἄχθος ἀρούρης “burden of the earth” (Il. 18.104; Od. 20.379).

23 For this analysis of ON fjǫrg, cf. Schaffner 2001, p. 193. NWGmc *ferg‑u ‘world, living beings’ reflects PGmc *fergw‑ō‑ and PIE *perḱu̯‑éh2‑, with NWGmc delabialisation between consonant and unstressed ‑u; for this development, cf. the NWGmc neut.pl. *sar‑u (OE searu ‘artifice, armor’, OHG saro ‘armor, gear’), the reflex of PGmc *sar‑wō‑ (Goth. sarwa ‘armor’), nom.pl. of the neut. *sar‑wa‑ ‘device, tool, weapon’ (Ringe, Taylor 2014, p. 16).

Auteur

Harvard University, Center for Hellenic Studies, Washington, D.C., USA

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search