Celtic Religions in the Roman Period: Personal, Local, and Global - Introduction

Anthony C King

This multi-authored book brings together new work, from a wide range of disciplinary vantages, on pre-Christian religion in the Celtic-speaking provinces of the Roman Empire. The chapters are the work of international experts in the fields of classics, ancient history, archaeology, and Celtic studies. It is fully illustrated with b&w and colour maps, site plans, photographs and drawings of ancient inscriptions and images of Romano-Celtic gods. The collection is based on the thirteenth workshop of the F.E.R.C.AN. project (fontes epigraphici religionum Celticarum antiquarum), which was held in 2014 in Lampeter, Wales. This paper is the general introduction to the volume, in a Welsh and English parallel text.

c e l t i c s t u d i e s p u b l i c a t i o n s XX Haeussler King CELTIC RELIGIONS IN THE ROMAN PERIOD Personal, Local, and Global celtic studies publications CSP–Cymru Cyf cover Celtic Religions Cocidius knockout terfynol.indd 1 celtic studies publications CSP–Cymru Cyf ISBN 978-1-891271-25-0 9 781891 271250 CELTIC RELIGIONS IN THE ROMAN PERIOD This multi-authored book brings together new work, from a wide range of disciplinary vantages, on pre-Christian religion in the Celtic-speaking provinces of the Roman Empire. The chapters are the work of international experts in the fields of classics, ancient history, archaeology, and Celtic studies. It is fully illustrated with b&w and colour maps, site plans, photographs and drawings of ancient inscriptions and images of Romano-Celtic gods. The collection is based on the thirteenth workshop of the F.E.R.C.AN. project ( fontes epigraphici religionum Celticarum antiquarum), which was held in 2014 in Lampeter, Wales. celtic studies publications edited by Ralph Haeussler & Anthony King CSP–Cymru Cyf 29/09/2017 16:03:37 Celtic Religions in the Roman Period Celtic Studies Publications xx Celtic Religions in the Roman Period Personal, Local, and Global edited by RALPH HAEUSSLER and ANTHONY KING Aberystwyth 2017 First published 2017 Copyright © CSP, Ralph Haeussler, Anthony King, and the individual authors All rights reserved. No part of this publication may be reproduced in any form or by any means, without permission from the Publisher Cover design by CSP-Cymru Cyf A Catalogue record for this book is available from the British Library. ISBN 978-1-891271-25-0 Celtic Studies Publications for customers in North America: Casemate Academic 1950 Lawrence Road Havertown, PA 19083 USA (phone: 1 610 853 9131) editorial correspondence: CSP-Cymru Cyf Centre for Advanced Welsh and Celtic Studies National Library of Wales Aberystwyth, Ceredigion SY23 3HH Wales TABLE OF CONTENTS Addresses of authors List of abbreviations v vii 1 Ralph Haeussler & Anthony King Crefyddau Celtaid yn y cyfnod Rhufeinig: personol, lleol a byd-eang — Celtic religion in the Roman period: personal, local, and global 2 John Koch & Fernando Fernández Palacios Some epigraphic comparanda bearing on the ‘pan-Celtic god’ Lugus 3 Jonathan Wooding Tyrannies of Distance? Medieval sources as evidence for indigenous Celtic and Romano-Celtic religion 1 4 Roger S. O. Tomlin A fourth-century ‘curse tablet’ from Uley 5 Daphne Nash Briggs Something old, something new: the names of Faunus in late Roman Thetford (Norfolk) and their Iron-Age background 6 Stephen Yeates The Roman religious landscape of Abingdon, Oxfordshire 7 Anthony King Carrying the Gods with them? Provenance and portability of altars to Romano-Celtic deities in Britain 8 Alessandra Esposito Talking to the gods: evidence for religious professionals and religious patterns in Roman Britain 9 Fernando Fernández Palacios The theonym *Conventina 71 Methodology 37 57 Britannia 79 103 119 151 165 Iberia 10 Patrizia de Bernardo Stempel Matres endeiterae, deus sanctus Endovelecos, dea Nave, and other indigenous and classical deities in the Iberian Peninsula 177 [ vi ] ta b le of co n te nt s 11 Blanca María Prósper Linguistic observations on two divinities of the Celtic Cantabri: 1) ERVDINO, divinity of the yearly cycle. 2) CABVNIAEGINO and the Celtic fate of IE *kap- and the Gaulish spindle whorl from Saint-Révérien 12 Silvia Alfayé Villa, Patrizia de Bernardo Stempel, Mª Cruz González Rodríguez, Manuel Ramírez Sánchez La diosa Du(v)itera en una inscripción de Tejeda de Tiétar (Cáceres) 13 Pilar Burillo-Cuadrato & Francisco Burillo-Mozota Looking at cosmology through the lens of hermeneutics and semanitcs 14 Manuela Alves Dias & Maria João Correia Santos The Gods that never were: new readings of the inscriptions of Penedo de Remeseiros (CIL II 2476), Penedo das Ninfas (CIL II 5607), Cueva del Valle (CIL II2.7 932) and Castro Daire (CIL II 5247) 207 229 253 273 Gallia & Belgica 15 Bernard Remy Les dieux au nom indigène et leurs cultores chez les Voconces de Vaison d’après les inscriptions 16 Florian Blanchard De Taranis au Jupiter cavalier à l’anguipède : ré lexions autour du substrat celtique dans la religion gallo-romaine. 17 Miranda Aldhouse-Green The Magician’s House. Druids, prayers and magic in Roman Gaul 18 Ralph Haeussler The importance of location: religious inscriptions from archaeological contexts 19 Audrey Ferlut Celtic goddesses from Gallia Belgica and the Germaniae: characterstics, dedicants, and ritual practices 287 309 325 339 363 Gallia Cisalpina 20 Cristina Girardi Sulle tracce dei luoghi di culto delle divinità plurali in Gallia Cisalpina 21 Paola Tomasi The cult of Hercules in central-eastern Transpadana (regio XI). Two case-studies from Laus Pompeia (Lodi Vecchio, Lodi) and Cedrate (Varese) 387 401 [ vii ] ta b le of con t e nt s Balkans & Danube Provinces 22 Alexander Falileyev Divine names from Latin inscriptions of Istria: some considerations 23 Marjeta Šašel Kos A sacred river landscape with a sanctuary. The worship of rivers in the south-eastern Alpine area 24 Vojislav Filipovic & Vladimir P. Petrovic Archaeological and epigraphic evidence for the Celtic Presence in the Upper Timachus River Valley (East Moesia Superior) 419 25 Hartmut Galsterer, Alfred Schäfer & Patrizia de Bernardo Stempel Mercurius Valdivahanus 26 Werner Petermandl CORPUS-F.E.R.C.AN. Germania Inferior: preliminary considerations and intentions 469 441 461 Germania Inferior Index of theonyms and epithets Topographic index 505 511 516 Cover photo: embossed silver plaque (height 11.3cm) discovered among 3rd-century rubbish at Bewcastle, the Roman Fanum Cocidi. Dedicated to a god bearing a Celtic name, Cocidius, the plaque depicts the god standing in a shrine with spear and shield (RIB 986, Tullie House Museum, Carlisle; photo: Anthony King). CONTRIBUTORS MIRANDA ALDHOUSE-GREEN Department of Archaeology, Cardiff University, Humanities Building, Colum Drive, Cardiff, CF10 3EU, Wales, aldhouse-greenmj@cardiff. ac.uk SILVIA ALFAYÉ VILLA Ciencias de la Antigüedad, Facultad de Filoso ía y Letras, Pedro Cerbuna, 12, Universidad Zaragoza, 50009 Zaragoza, Spain, silvia.alfaye@gmail.com MANUELA ALVES DIAS Centro de Estudos Clássicos, Universidade de Lisboa, Lisboa, Portugal, epigraphica@gmail. com FLORIAN BLANCHARD Centre de Recherche Bretonne et Celtique (CRBC), Université de Bretagne Occidentale, Faculté des Lettres et Sciences Humaines, 20 Rue Duquesne – CS 93837, 29238 Brest Cedex 3, France, lorian3.blanchard@orange.fr FRANCISCO BURILLO-MOZOTA Facultad de Ciencias Sociales y Humanas, Universidad de Zaragoza, Teruel, Spain, burillo@unizar.es Mª PILAR BURILLO-CUADRADO Ciencias de la Antigüedad, Facultad de Filoso ía y Letras, Pedro Cerbuna, 12, Universidad Zaragoza, 50009 Zaragoza, Spain, mpilar.burillo@gmail.com MARIA JOÃO CORREIA SANTOS Centre of Classic Studies (Centro de Estudos Clássicos), University of Lisbon, Portugal, mj.correiasantos@letras.ulisboa.pt PATRIZIA DE BERNARDO STEMPEL Departamento de Estudios Clásicos, Universidad del País Vasco / Euskal Herriko Unibertsitatea, Vitoria / Gasteiz, Spain, patrizia.debernardo@ehu.es ALESSANDRA ESPOSITO Classics Department, King’s College London, Con tr i bu tor s Strand, London WC2R 2LS, alessandra.esposito@kcl.ac.uk England, ALEXANDER FALILEYEV Aberystwyth, Wales, a.falileyev@gmail.com FERNANDO FERNÁNDEZ PALACIOS Centre for Advanced Welsh and Celtic Studies, University of Wales, Aberystwyth, SY23 3HH, Wales, mbuchanscot@gmail.com AUDREY FERLUT Université Jean Moulin Lyon 3, UMR 5189, 69007 Lyon, France, aferlut1@ac-lyon.fr VOJISLAV FILIPOVIC Institute of Archaeology, Belgrade, Serbia, v ilipov1@gmail.com HARTMUT GALSTERER Institut für Geschichtswissenschaft, Abteilung für Alte Geschichte, Universität Bonn, Bonn, Germany CRISTINA GIRADI Università degli Studi di Padova, DISSGEA, Piazza Capitaniato 7, 35100 Padova, cristina. girardi@yahoo.it Mª CRUZ GONZÁLEZ RODRÍGUEZ Estudios Clásicos, Universidad del País Vasco/ Euskal Herriko Unibertsitatea, Vitoria/ Gasteiz, Spain, cruz.gonzalez@ehu.es RALPH HAEUSSLER Faculty of Humanities, University of Wales Trinity Saint David, Lampeter, SA48 7ED, Wales, ralph.haussler@uclmail.net ANTHONY KING Department of Archaeology, University of Winchester, Sparkford Road, Winchester SO22 4NR, England, tony.king@winchester.ac.uk JOHN KOCH Centre for Advanced Welsh and Celtic Studies, [ ix ] University of Wales, Aberystwyth, SY23 3HH, Wales, jtk@wales.ac.uk DAPHNE NASH BRIGGS Formerly School of Archaeology, Oxford University, England, d.briggs@classics.oxon.org WERNER PETERMANDL Institut für Alte Geschichte und Altertumskunde, Karl-Franzens-Universität Graz, Universitätsplatz 3/II, 8010 Graz, Austria, werner.petermandl@uni-graz.at VLADIMIR P. PETROVIC Institut des études balkaniques de l`Académie serbe des sciences et des arts, Knez Mihailova 35/IV, 11000 Belgrade, Serbia, vladimir. arheolog@gmail.com BLANCA MARÍA PRÓSPER Departamento de Filología Clásica e Indoeuropeo, Universidad de Salamanca, Plaza de Anaya s/n, 37008 Salamanca, Spain, indoling@usal.es MANUEL RAMÍREZ SÁNCHEZ Universidad de Las Palmas de Gran Canaria, Juan de Quesada, 30, 35001 Las Palmas de Gran Canaria, Spain, manuel.ramirez@ulpgc.es BERNARD REMY Laboratoire Universitaire Histoire Cultures Italie Europe, Grenoble (LUHCIE) Centre Camille Jullian, Maison méditerranéenne des sciences de l'homme, 5, rue du Château de l’horloge, BP 647, 13094 Aix-en-Provence, France, bernard.remy07@orange.fr MARJETA ŠASEL KOS Institute of Archaeology, ZRC-SAZU, Novi trg 2, 1000 Ljubljana, Slovenia, mkos@zrc-sazu.si ALFRED SCHÄFER Römisch-Germanisches Museum der Stadt Köln, Köln, Germany, alfred.schaefer@stadtkoeln.de [x] Cont ri bu tor s PAOLA TOMASI Dipartimento di Studi Umanistici, Università degli Studi di Pavia, 27100 Pavia, Italy, paola. tomasi@unipv.it ROGER S. O. TOMLIN Wolfson College, Oxford University, Oxford, OX2 6UD, England, roger.tomlin@wolfson. ox.ac.uk JONATHAN WOODING School of Literature, Arts and Media, University of Sydney, NSW 2006, Australia, jonathan.wooding@sydney.edu.au STEPHEN YEATES Wolfson College, Oxford, Oxfordshire, OX2 6UD, England, stephenyeates@pobroadband. co.uk ABBREVIATIONS AALR = Atlas antroponímico de la Lusitania romana, Mérida & Burdeos, Ausonius Éditions, 2003. ABC = Cottam, E., P. de Jersey, C. Rudd & J. Sills 2010 Ancient British Coins. Aylsham, Chris Rudd. AC = Annales Cambriae, cited from J. Morris (ed. & trans.), British History and the Welsh Annals/ Nennius, Arthurian Period Sources 8. London, Phillimore 1980. AE = L’Année épigraphique. Revue des publications épigraphiques relatives à l’Antiquité romaine. Paris 1888–. AEW = de Vries, J. 2004 Altnordisches etymologisches Wörterbuch. Leiden. AW = Schützeichel, R. 2005 Althochdeutsches Wörterbuch. Tübingen. AIJ = Hof iller, V. & B. Saria 1938 Antike Inschriften aus Jugoslawien. 1: Noricum und Pannonia superior (re-print Amsterdam 1970). ANRW = Aufstieg und Niedergang der Römischen Welt. Berlin & New York: Walter de Gruyter 1972–. Caes. BGall. = Caesar, de bello Gallico CAG = Carte archéologique de la Gaule. Les Alpesde-Haute-Provence. Paris, Académie des Inscriptions et Belles Lettres. CCelt. = Continental Celtic. CIIC = Macalister, R. A. S. 1996 Corpus Inscriptionum Insularum Celticarum. Blackrock, Co. Dublin, Four Courts Press. CIL = Corpus Inscriptionum Latinarum. Berlin. 1861–. CILCC I = Esteban Ortega, J. 2007 Corpus de inscripciones latinas de Cáceres. I. Norba, Universidad de Extremadura, Cáceres. CIRG I = Corpus de inscripcións romanas de Galicia. Santiago de Compostela, Consello da Cultura Galega: vol. I, Provincia de A Coruña (1991), vol. II, Provincia de Pontevedra (1994). CPILC = Hurtado de San Antonio, R. 1977 Corpus Provincial de Inscripciones Latinas de Cáceres. Cáceres. CIRP Salamanca = Alonso Ávila, Á & S. Crespo Ortiz de Zárate 1999 Corpus de inscripciones romanas de la provincia de Salamanca. Valladolid. CPILC = Hurtado de San Antonio, R. 1977 Corpus provincial de inscripciones Latinas de Cáceres, Cáceres. Cáceres, Diputación Provincial. Servicios Culturales. CR = The Cartulary or Redon, ed. A. de Courson. Paris, 1863. CSIR = Corpus Signorum Imperii Romani. Corpus of sculpture of the Roman world, 18 vols. London or Oxford or Berlin. 1975–2004. dat. = dative. DCC/DCCPlN = Falileyev, A. (in collaboration with A. E. Gohil & N. Ward) 2010 Dictionary of Continental Celtic Place-Names. Aberystwyth, CMCS. DE = Dizionario epigra ico di antichità romane. Roma : Istituto Italiano per la Storia antica, 1895–1997. DIL = (Contributions to a) Dictionary of the Irish Language. Dublin, Royal Irish Academy, 1913– 1976. DN = divine name. EDCS = Epigraphik-Datenbank Clauss-Slaby http://www.manfredclauss.de. EDR = Epigraphic Database Roma. EE = Ephemeris epigraphica. ERAv = Hernando Sobrino, Mª R. 2005 Epigra ía romana de Ávila, Petrae Hispaniarum, 3/4, Burdeos & Madrid, Ausonius Éditions. ERCanosa = Grelle, F. & M. Pani 1990 Le Epigra ia Romane di Canosa. Bari. ERPL = Rabanal Alonso, M. A., & S. M.ª García Martínez 2001 Epigra ía romana de la provincia de León: revisión y actualizaciones. León. ERSegovia = Santos Yanguas J. 2005 Epigra ía romana de Segovia y su provincia. Segovia. ERZamora = Bragado Toranzo, J. M.a 1991 Fuentes literarias y epigrá icas de la provincial de Zamora y su relación con las vías romanas de la Cuenca del Duero. Universidad de León. Esp. = Espérandieu, E. 1910–1928, Recueil général des Bas-reliefs, statues et bustes de la Gaule romaine, vols I-X. Paris, Editions Ernest Leroux; Espérandieu, E. & Lantier, R. 1938–1966 Recueil général des bas-reliefs, statues et bustes de la Gaule romaine,vols XI-XV Paris, Editions Ernest Leroux. Espérandieu, E. 1931 Recueil général des basreliefs, statues et bustes de Germanie. Paris, Editions Ernest Leroux. EWD = Pfeifer, W. 2003 Etymologisches Wörterbuch des Deutschen. München (1st edition: Berlin, Deutscher Taschenbuch Verlag 1989). [ xii ] A bb r e v i ation s FE = Ficheiro Epigrá ico. Suplemento de Conimbriga. Coimbra. FN = family name. GED = Lehmann, W. P. 1986 A Gothic Etymological Dictionary. Based on the 3rd edition of Vergleichendes Wörterbuch der Gotischen Sprache by Sigmund Feist. Leiden. HAE = Hispania Antiqua Epigraphica. Suplemento anual de Archivo Español de Arqueología. Instituto de Arqueología y Prehistoria «Rodrigo Caro». Madrid, Consejo Superior de Investigaciones Cientí icas. HB = The Historia Brittonum of ‘Nennius’, cited from Monumenta Germaniae Historica, Chronica Minora, ed. Th. Mommsen. Berlin 1898; also, British History and the Welsh Annals/Nennius, Arthurian Period Sources 8, ed. & trans. J. Morris. London, Phillimore, 1980; The Historia Brittonum 3: The ‘Vatican’ Recension, ed. D. N. Dumville. Cambridge, Brewer, 1985. HD = Heidelberg epigraphische Datenbank, http:// edh-www.adw.uni-heidelberg.de/home HEp = Hispania Epigraphica. Revista del Archivo Epigrá ico de Hispania. Madrid, Universidad Complutense. Online: http://eda-bea.es/. Holder = Holder, A. 1896–1910 Alt-celtischer Sprachschatz. 3 vols. Leipzig, B. G. Teubner (reprint 1961–1962, Graz, Akademische Druckund Verlagsanstalt). ICelt. = Insular Celtic. IE = Indo European. IEW = Pokorny, J. 1959–1969 Indogermanisches etymologisches Wörterbuch. 2 vols. Bern, Francke. ILA-Bordeaux = Inscriptions Latines d’Aquitaine: Bordeaux, ed. L. Maurin, M. Navarro Caballero, D. Barraud, C. Brial & A. Zieglé, Bordeaux/Paris 2010. ILER = Vives, J. 1971–1972 Inscripciones latinas de la España romana. Barcelona. ILGN = Espérandieu, É. 1929 Inscriptions Latines de Gaule Narbonnaise. Paris, E. Leroux. ILJug = Šašel, A. & J., Inscriptiones latinae quae in Iugoslavia inter annos MCMXL et MCMLX repertae et editae sunt (= Situla 5). Ljubljana 1963; Šašel, A. & J., Inscriptiones latinae quae in Iugoslavia inter annos MCMLX et MCMLXX repertae et editae sunt (= Situla 19). Ljubljana 1978; Šašel, A. & J., Inscriptiones Latinae quae in Iugoslavia inter annos MCMII et MCMXL repertae et editae sunt (= Situla 25). Ljubljana 1986. ILLPRON = Inscriptionum Lapidarium Latinarum Provinciae Norici usque ad annum MCMLXXXIV repertarum indices. Berlin 1986. ILN = Inscriptions Latines de Narbonnaise. Paris, CNRS éditions (ILN-Die = Rémy, B. et al. 2012, VII, Voconces. VII, 1, Die; ILN-Vienne = Rémy, B. 2004–2005 V. Vienne, 3 vols). ILS = Dessau, H. 1892–1916 Inscriptiones Latinae Selectae, 3 vols. Berlin. ILSl = Inscriptiones Latinae Sloveniae, 1: M. Lovenjak, Neviodunum. Ljubljana 1998. Inscr. Aqu. = J.B. Brusin, J. B. 1991–1993 Inscriptiones Aquileiae, 3 vols, Udine. Inscr.It. = Inscriptiones Italiae, Rome 1931–. IRC I = Fabre, G., M. Mayer & I. Rodà, eds 1984 Inscriptions romaines de Catalogne, I, Barcelone (sauf Barcino). Paris, Diffusion de Boccard. IRG = Bouza Brey, F. & A. D’Ors 1949 Inscripciones romanas de Galicia. Santiago de Compostela, Instituto Padre Sarmiento de Estudios Gallegos 1949–. IRLugo = Arias, F., P. Le Roux & A. Tranoy 1979 Inscriptions romaines de la province de Lugo. Paris, Diffusion de Boccard. IRMN = Castillo, C., J. Gómez-Pantoja & Mª D. Mauleón 1981 Inscripciones romanas del Museo de Navarra. Pamplona, Diputación de Navarra. IRPP = Hernández Guerra, L. 1994 Inscripciones romanas en la provincia de Palencia. Valladolid. JP = catalogue numbers from Johns, C. M. & T. Potter 1983 The Thetford Treasure. London, BMP. LEIA = Vendryes, J., Bachellery, É. & Lambert, P.Y. 1959– Lexique étymologique de l'irlandais ancien. Dublin & Paris, CNRS Éditions. LHEB = Jackson, K. H. 1994 Language and History in Early Britain: A Chronological Survey of the Brittonic Languages from the 1st to the 12th c. AD. 2nd rev. ed. Dublin, Four Courts Press ( irst published by Edinburgh University Press 1953). LKA = Sievers, S., O. H. Urban& P. C. Ramsl 2012 Lexikon zur keltischen Archäologie, 2 vols. Wien, Österreichische Akademie der Wissenschaften (Mitteilungen der Prähist. Kommission 73). Lib.Lan. = The Text of the Book of Llan Dâv: Reproduced from the Gwysaney Manuscript, ed. J. G. Evans, J. Rhŷs, Facsimile reprint, Aberystwyth, National Library of Wales, 1979; irst published, Oxford, J. G. Evans, 1893; datings of the personal names in Lib.Lan. are from W. Davies, Llandaff Charters. Aberystwyth, National Library of Wales 1979. LIV = Rix, H. et alii eds 11998/22001 Lexikon der Indogermanischen Verben. Die Wurzel und ihre Primärstammbildungen. Wiesbaden, Reichert. A bbr e v i ation s LS = Lewis, C. T. & C. Short 1955 A Latin Dictionary. Oxford (original edition, 1879). lupa = http://lupa.at/ MLH IV = Untermann, J. (with D. S. Wodtko) 1997 Monumenta Linguarum Hispanicarum. IV. Die tartessischen, keltiberischen und lusitanischen Inschriften. Wiesbaden, Ludwig Reichert. M.R.M. = Morphological Residual Model. NPC = Delamarre, X. 2007 Noms de personnes celtiques dans l’épigraphie classique. Paris, Errance. NTS = Hartley, B. R. and Dickinson, B. M. , eds 2008–2012 Names on Terra sigillata, An Index of Makers’ Stamps & Signatures on Gallo-Roman Terra sigillata (Samian Ware), 9 vols. London, Bulletin of the Institute of Classical Studies Supplement 102. NWÄI = De Bernardo Stempel, P. 1999 Nominale Wortbildung des älteren Irischen: Stammbildung und Derivation, Tübingen, Max Niemeyer/ W. de Gruyter. OBret. = Old Breton. OEng. = Old English. OHG =Old High German. OIc. = Old Icelandic. OIr = Old Irish. OPEL = Onomasticon Provinciarum Europae Latinarum, ed. Lőrincz, B. & Redő, F. ex materia ab Mócsy, A., Feldmann, R., Marton E. et Szilágyi, M. collecta. 4 vols. Budapest, Archaeolingua, 1994 & Wien, Forschungsgesellschaft Wiener Stadtarchäologie, 1999–2002. PAS = Portable Antiquaties Scheme. PCelt. = Proto-Celtic. P.Dura = Bradford Welles, C. et alii 1959 The Excavations at Dura-Europos. Final Report, v, part I, The Parchments and Papyri. New Haven. PIE = Proto Indo-European. pl. = plural. Plin. HN = Plinius, Historia Naturalis. PlN = place name. PN = personal name. Pokorny, IEW = Pokorny, J. 1959 Indogermanisches Etymologisches Wörterbuch. [ xiii ] Heidelberg. RAP = Garcia, J. M. 1991 Religiões antigas de Portugal. Aditamentos e observações às «Religiões da Lusitânia» de J. Leite de Vasconcelos. Fontes epigrá icas. Lisboa. RE = Paulys Realencyclopädie der classischen Altertumswissenschaft. 84 vols. Berlin 1894– 1978. RIB = The Roman Inscriptions of Britain, 3 vols. Oxford and Stroud. RIG = Recueil des inscriptions gauloises. Paris, CNRS. I. Inscriptions gallo-grecques (1985), II.1 Textes gallo-étrusques, Textes gallo-latins sur pierre (1988), II.2 Textes gallo-romains sur instrumentum (2002). RINMS = M. Šašel Kos 1997 The Roman Inscriptions in the National Museum of Slovenia, Ljubljana. RSK = B. & H. Galsterer 1975 Die römischen Steininschriften aus Köln. Köln, RömischGermanisches Museum. SI = Pais, H. 1884 Corporis Inscriptionum Latinarum Supplementa Italica, I. Additamenta ad vol. V Galliae Cisalpinae, ex typis Salviucci. Roma. Supp.It. = Supplementa Italica. Rome 1981–. Tab. Peut. = Tabula Peutingeriana. Tab. Sulis = Tomlin, R. S. O. 1988 ‘The curse tablets‘, in B. Cunliffe (ed.), The temple of Sulis Minerva at Bath, vol. 2. The inds from the sacred spring. Oxford, Oxford University Committee for Archaeology, Monograph 16, 1988, 4–277. Tac. ann. = Tacitus, Annales. ThesCRA = Thesaurus Cultus et Rituum Antiquorum. Los Angeles, The J. Paul Getty Museum. 5 vols. 2004–2005. ThLL/TLL = Thesaurus Linguae Latinae. TVindol = Bowman, A. K. & J. D. Thomas, The Vindolanda writing-tablets (Tabulae Vindolandeses), 4 vols, London 1994– (online: http://vindolanda.csad.ox.ac.uk/). u.a. = unter anderem. I CREFYDDAU CELTAIDD YN Y CYFNOD RHUFEINIG: PERSONOL, LLEOL A BYD-EANG CELTIC RELIGIONS IN THE ROMAN PERIOD: PERSONAL, LOCAL, AND GLOBAL Ralph Haeussler & Anthony King “The inhabitants (of Gaul) are proud and superstitious and once they were that barbaric that they considered man as the best victim and most agreeable to the gods. There are remnants of these savage customs that are today abolished (...)” (Pomponius Mela, De Chorographia III 2, 18–9) “Mae trigolion (Gâl) yn falch ac ofergoelus ac unwaith roeddent mor farbaraidd nes eu bod yn ystyried mai pobl oedd yr ysglyfaeth orau a’r un fwyaf wrth fodd y duwiau. Mae gweddillion yr arferion barbaraidd hyn sydd wedi eu dileu erbyn heddiw (...)” (Pomponius Mela, De Chorographia III 2, 18–9) F .E.R.C.AN.’s aim is to compile a corpus of the fontes epigraphici religionum Celticarum antiquarum, the epigraphic sources for the ancient (i.e. pre-medieval) Celtic religions.1 The number of pre-Roman epigraphic sources is extremely limited, as most societies did not consider it appropriate to use writing 1 For post-antiquity ‘Celtic’ inscriptions from AD 400–1000, cf. the Celtic Inscribed Stones Database: http://www.ucl.ac.uk/archaeology/cisp/ database/ – for the F.E.R.C.AN. project, cf. e.g., Hainzmann & De Bernardo Stempel 2013 and the paper by W. Petermandl in this volume. N 1 od F.E.R.C.AN. yw llunio corpws o’r fontes epigraphici religionum Celticarum antiquarum, sef y ffynonellau epigraf ig ar gyfer crefyddau Celtaidd yr hen fyd (h.y. cyn yr Oesoedd Canol).1 Mae nifer y ffynonellau epigraf ig cynRufeinig yn gyfyngedig dros ben, gan nad oedd y rhan fwyaf o gymdeithasau’n Ar gyfer arysgrifau ‘Celtaidd’ ar ôl cyfnod yr Hen Fyd rhwng OC 400–1000, cf. Celtic Inscribed Stones Database: http://www.ucl.ac.uk/ archaeology/cisp/database/ – ar gyfer y prosiect F.E.R.C.AN., cf. e.e., Hainzmann a De Bernardo Stempel 2013 a phapur W. Petermandl yn y gyfrol hon. [2] 2 3 I. c e lt ic r e l igions in th e Rom a n pe r io d for religious matters in the Iron Age.2 Most of the evidence for this project therefore primarily comes from the Roman period. But what can Roman inscriptions really tell us about ‘Celtic’ religions, if anything? Obviously, the project presumes a certain persistence of Iron Age cults into and throughout the Roman period, which makes sense since the Romans generally did not impose their own cults on the conquered and provincial populations.3 As we are told by Caesar BGall. 6.14. There are some pre-Roman inscriptions, written in Celtic languages (notably Lepontic/Gallo-Etruscan, Gallo-Greek, and Celtiberian inscriptions), but only few refer to the religious sphere. For 2 example, there is the famous bilingual LatinCeltic inscription from Vercelli/Vercellae in Cisalpine Gaul (1st century BC), written in socalled Lepontic script (RIG II.1, E-2). As is common with these early inscriptions, they do not mention any names of deities, but primarily the names of dedicants (also cf. e.g. the dedication from Briona: RIG II.1, E-1). In the case of Vercelli, the inscription does provide some details of the cult: the phrase campus …. deis et hominbus … dedit refers to a ‘campus … given … to gods and humans’ (though P. de Bernardo Stempel 2011 suggests that deis et hominibus – a Latin rendering of the Celtic teuoχtom – was primarily a legal phrase). In the south of Gaul, we ind some Gallo-Greek inscriptions that mention deities (usually not before the 1st century BC), notably Belenos (RIG G-28; CAG 13/1 100; CAG 13/1 55; Lejeune, 1968–1969, 61–67, no 20), Belisama (RIG G-153), Matrebo Glanaikabo (RIG G-64), Matrebo Namausikabo (RIG G-203), Roklosia (RIG G-65), Taranoou (Taranis? – RIG G-27), and Larasso (IGF 207; but for De Bernardo Stempel 2007, 70 an Aquitanian theonym). And in Iberia we ind Celtiberian inscriptions mentioning inter alia the possible theonyms Neto/Neito and Tokoitos (cf. Abascal 2002). For aspects of continuity and persistence in provincial cults, cf. papers in Haeussler and King (eds) 2007 and 2008). Cf. Festus, s.v. municipalia sacra 146L: this passage shows that 3 the Roman ponti ices, the highest priests of the Roman state, wanted people to worship their ancestral cults, even in the context of a Romancitizen community. ystyried ei bod yn briodol defnyddio ysgrifennu ar gyfer materion crefyddol yn yr Oes Haearn.2 Daw’r rhan fwyaf o’r dystiolaeth ar gyfer prosiect hwn, felly, o’r cyfnod Rhufeinig. Ond beth, os unrhyw beth, y gall arysgrifau Rhufeinig ei ddweud wrthym am y grefydd ‘Geltaidd’? Yn amlwg, rhagdybia’r prosiect fod defodau’r Oes Haearn wedi rhyw barhau i’r cyfnod Rhufeinig a thrwy gydol y cyfnod hwn, sy’n gwneud synnwyr gan nad oedd y Rhufeiniaid, fel arfer, yn gorfodi eu defodau eu hunain ar bobloedd orchfygedig yn y taleithiau.3 A oes Fel yr adroddir wrthym gan Gesar, BGall. 6.14. Mae rhai arysgrifau cyn-Rufeinig wedi eu hysgrifennu mewn ieithoedd Celtaidd (yn enwedig arysgrifau Lepontig/Galaidd-Etrwscaidd, Galaidd-Roegaidd, a Cheltiberaidd), ond ychydig iawn yn y cylch crefyddol. Er enghraifft, mae arysgrifen ddwyieithog Ladin a Chelteg o Vercelli/Vercellae yng Ngâl Isalpaidd (y ganrif 1af CC), wedi ei hysgrifennu yn yr ysgrifen Lepontig fel y’i gelwir (RIG II.1, E-2). Fel sy’n gyffredin gyda’r arysgrifau hyn, nid ydynt yn crybwyll unrhyw enwau duwiau, ond yn bennaf enwau offrymwyr (hefyd cf. e.e. y cysegriad o Briona: RIG II.1, E-1). Yn achos Vercelli, mae’r ysgrifen yn rhoi manylion y ddefod: cyfeiria’r ymadrodd campus …. deis et hominbus … dedit at ‘campus … a roddwyd … i dduwiau a phobl’ (er yr awgryma P. de Bernardo Stempel 2011 mai term cyfreithiol yn anad dim oedd deis et hominibus – ffurf Ladin ar y teuoχtom Celtaidd). Yn ne Gâl, ceir rhai arysgrifau Galaidd-Roegaidd sy’n sôn am dduwiau (fel arfer, nid cyn y ganrif 1af CC), yn enwedig Belenos (RIG G-28; CAG 13/1 100; CAG 13/1 55; Lejeune, 1968–1969, 61–67, no 20), Belisama (RIG G-153), Matrebo Glanaikabo (RIG G-64), Matrebo Namausikabo (RIG G-203), Roklosia (RIG G-65), Taranoou (Taranis? – RIG G-27), a Larasso (IGF 207ar gyfer De Bernardo Stempel 2007, 70 enw duw o Acwitania). Ac yn Iberia cawn arysgrifau Celtiberiaidd sy’n sôn inter alia am yr enwau duwiau Neto/Neito a Tokoitos (cf. Abascal 2002). Am agweddau ar barhad a dygnwch mewn defodau yn y taleithiau, cf. papurau yn Haeussler a King (gol) 2007 a 2008). Cf. Festus, s.v. municipalia sacra 146L: dengys y darn hwn fod y ponti ices, sef offeiriaid uchaf y wladwriaeth Hae ussle r & King 4 5 6 7 [ 3 ] Can we assess how much of these Celtic ‘superstitions’ and ‘savage customs’, as Pomponius Mela pejoratively called them around A.D. 43, survived into the Roman period, and in which form? What exactly was ‘abolished’,4 and what are Pomponius Mela’s ‘remnants’?5 This also prompts the question whether the Roman conquest initiated a rupture in religious understandings, for instance, if we presume that the druids’ alleged eradication had an impact on religious beliefs across the Keltiké.6 It is also possible that local cults merely adapted gradually to new social needs, new cultural conventions and new media (i.e. writing and sculpture) in a changing world, and thus that there was some persistence of autochthonous religious understandings and practices that Romans pejoratively called ‘superstitions’. As we shall see, in most cases one can identify combinations of persistence, transformation, innovation, and rupture, across the Roman West to varying degrees. modd i ni asesu faint o’r ‘ofergoelion’ ac ‘arferion barbaraidd’ Celtaidd hyn, a defnyddio enw sarhaus Pomponia Mela arnynt tua 43 OC, a oroesodd i’r cyfnod Rhufeinig, ac ar ba ffurf? Beth yn union a ‘ddilëwyd’,4 a beth yw ‘gweddillion’ Pomponius Mela?5 Mae hyn hefyd yn codi’r cwestiwn a wnaeth y goncwest Rufeinig greu hollt o ran dealltwriaeth grefyddol, er enghraifft, a rhagdybio i ddilead honedig y derwyddon effeithio ar gredoau crefyddol ar draws y Keltiké.6 Mae hefyd yn bosibl i ffur iau lleol ar addoli gael eu haddasu yn unig i anghenion cymdeithasol newydd, confensiynau diwylliannol newydd a’r cyfryngau newydd (h.y. ysgrifennu a cher lunio) mewn byd a oedd yn newid, ac o’r herwydd glynodd rhyw ddealltwriaeth o arferion crefyddol brodorol a ddisgri id yn sarhaus yn ‘ofergoelion’ gan y Rhufeiniaid. Yn y rhan fwyaf o achosion, fel y gwelwn, gellir nodi cyfuniadau o barhau, trawsnewid, arloesi a rhwygo, ar draws y Gorllewin Rhufeinig i wahanol raddau. The problematic nature of the evidence for ‘Celtic religions’ Natur broblemus y dystiolaeth ynghylch ‘crefyddau Celtaidd’ Interestingly, for Julius Caesar the Galli had more or less the same beliefs of the gods as the other peoples.7 Does Yn ddiddorol, i Iwl Cesar roedd gan y Galli yr un credoau am y duwiau Perhaps just the alleged human sacri ices, whose existence in the late Iron Age can be disputed. Also cf. Pliny HN 30.4 on the rituals that were 4 said to have been abolished: magical art, monstrous rites, and human sacri ice. 5 On the druids’ last stand on Anglesey/Mona, cf. Tac. Ann. 14.29–30. Caes. BGall. 6.17: de his eandem fere, quam 6 reliquae gentes, habent opinionem. But we am i bobl addoli duwiau eu hyna iaid, hyd yn oed yng nghyd-destun cymuned o ddinasyddion Rhufeinig.. O bosibl yr aberthau dynol honedig, y gellir herio iddynt ddigwydd o ddiwedd yr Oes. Hefyd cf. Pliny HN 30.4 ar y defodau y dywedwyd iddynt gael eu dileu: celf hudol, defodau angen ilaidd ac aberthau dynol. Ar sa iad olaf y derwyddon ar Ynys Môn/Mona, cf. Tac. Ann. 14.29–30. [4] I. c e lt ic r e l igions in th e Rom a n pe r io d his famous list of the Gaul’s main gods – Mercury, Apollo, Mars, Jupiter, and Minerva, as well as Dispater – already lead the way to the local ‘panthea’ of the Roman period (if we assumed for the sake of argument that his accounts were reliable)?8 Who is Mercury in Roman Gaul, Britain or Spain: was he ever the Roman god, adopted by the natives or imported by soldiers and colonists, or was this not predominantly a Latin name for an indigenous god whose ‘real’ name escapes us?9 And is Caesar’s Apollo 8 9 also should take into account that, despite his 7 rather generalizing statements, he also asserts that the Gauls differ by the language, institutions, and laws: hi omnes lingua, institutis, legibus inter se differunt (Caes. BGall. 1.1.2). Caes. BGall. 6.17.1–2; 6.18.1; for general source criticism, cf. Haeussler 2014 with further bibliography. Also cf. Parthenios of Nikaia, in 8 Zwicker 1935, vol. 2, 20 for whom Hercules, not Dispater, was the ancestor of all Gauls, which may explain Hercules’ popularity across the Keltiké. Obviously, if we presume the Graeco-Roman concept of polis religion or civic religion, we can hardly assume that anybody in the an- 9 cient world would have made a dedication to a Roman Mercury outside Rome or Italy, but even a Roman soldier or magistrate might have worshipped a local ‘incarnation’ of Mercury (or any other deity). It seems likely that these local Mercuries were becoming increasingly similar across the empire, due to the increasing spatial mobility and exchange of ideas, while taking up more and more substance from their Graeco-Roman counterparts (due to the spread of Graeco-Roman literature, myth, and art); at the same time, we also see the local Mercury with a purse, con irming Caesar’s statement that he was considered ‘the one to have the greatest strength for the proﬁt of money and trade’, hunc ad quaestus pecuniae mercaturasque habere vim maximam (…) (Caes. BGall. 6.17). And occasionally we ind other, non-Classical attributes, such as a tree, like on the votive pillar from Caveirac (Gard) combining Mercury’s Classical attributes in a novel way: CIL XII 3090 = 4136 = Esp. I 441. But can we associate these types of ‘Romano-Celtic’ Mercuries with any pre- â’r bobloedd eraill fwy neu lai.7 A yw ei restr enwog o brif dduwiau Gâl – Mercher, Apolon, Mawrth, Iau a Minerva, yn ogystal â Dispater – eisoes yn arwain y ffordd at ‘panthea’ lleol y cyfnod Rhufeinig (a thybio am y tro fod ei adroddiadau’n ddibynadwy)?8 Pwy oedd Mercher yng Ngâl, Prydain neu Sbaen y Rhufeiniaid: a fu erioed yn dduw Rhufeinig, a fabwysiadwyd gan y brodorion neu a fewnforiwyd gan ilwyr a gwladychwyr, neu ai enw Lladin oedd hwn yn bennaf ar gyfer duw brodorol nad yw ei enw ‘go iawn’ yn hysbys?9 Ac a yw Apolon Cesar yn Caes. BGall. 6.17: de his eandem fere, quam reliquae gentes, habent opinionem. Ond er gwaethaf ei sylwadau cyffredinol, dylem gymryd i ystyriaeth ei fod hefyd yn dweud bod y Galiaid yn amrywio o ran iaith, sefydliadau, a chyfreithiau: hi omnes lingua, institutis, legibus inter se differunt (Caes. BGall. 1.1.2). Caes. BGall. 6.17.1–2; 6.18.1; am feirniadaeth ﬀynhonnell gyﬀredinol, cf. Haeussler 2014 gyda llyfryddiaeth bellach. Hefyd cf. Parthenios o Nikaia, yn Zwicker 1935, cyf. 2, 20 a gredai mai Ercwlff, yn hytrach na Dispater, oedd hyna iad yr holl Galiaid, a gallai hyn esbonio poblogrwydd Ercwlff ar draws y Keltiké. Os rhagdybir mai’r cysyniad Groegaidd-Rufeinig o grefydd y polisi neu grefydd ddinesig, mae’n amlwg mai go brin y gellir rhagdybio y byddai unrhyw un yn yr Hen Fyd wedi cysegru i Mercher y Rhufeiniaid y tu allan i’r Eidal, ond gallai hyd yn oed milwr neu ynad Rhufeinig wedi addoli ‘ymgnawdoliad’ lleol o Fercher (neu unrhyw dduw arall). Wrth gwrs, gallwn dybio bod y duwiau Mercher hyn yn ymdebygu’n fwy ac yn fwy ar draws yr ymerodraeth, oherwydd mwy o symudedd gofodol a chyfnewid syniadau, gan fabwysiadu mwy a mwy o sylwedd yr elfennau Groegaidd-Rufeinig cyfatebol (yn sgil lledaeniad llenyddiaeth, chwedlau a chelfyddyd Roegaidd-Rufeinig); ar yr un pryd, gwelir hefyd y Mercher lleol gyda phwrs, gan gadarnhau datganiad Cesar yr ystyrid bod ganddo’r nerth gorau ar er lles arian a masnach’, hunc ad quaestus pecuniae mercaturasque habere vim maximam (…) (Caes. BGall. 6.17). O bryd i’w gilydd, ceir priodoleddau eraill, nad ydynt yn rhai Clasurol , megis coeden, fel yn achos y piler addunedol o Caveirac (Gard) sy’n cyfuno priodoleddau Clasurol Mercher mewn modd Hae ussle r & King based on his personal knowledge of the Celtic Belenos and Grannus?10 And who is Caesar’s Minerva in Gaul, considering that other goddesses, like Juno, Maia/Rosmerta and the various ‘mothers’, are more widely attested?11 This raises inter alia two important questions: Caesar’s interpretationes reveal the problem of translatability; can there ever be two identical deities in two different cultures or religions? No. Consequently, different people would have made different identi ications between Celtic, Greek and Roman deities depending on their personal knowledge of the local cult or deity, on their personal interpretation, and depending on time, since a cult’s meaning is bound to change over time. How reliable are Caesar’s generalising accounts? Local people made their personal interpretatio in the Roman empire, frequently diverging from Caesar’s interpretation: for example, 10 11 Roman Celtic deity (or are we dealing with a more paci ied version suitable for the Roman Principate?). The Berne Scholies on Lucan’s Pharsalia associate Mercury (and also Mars!) with both Esus and Toutatis (Zwicker 1935, I, 51-2; cf. Duval 1989 for a discussion of ancient sources for Esus, Toutatis, and Taranis). But this raises once again the question whether these Celtic words refer to a god, or whether they are merely used as a surname or epithet with esus merely meaning ‘god’ (cf. Meid 2003) and toutatis meaning ‘of the people’ (Delamarre 2003: 294-5). Apollo is actually rather rare across the Roman 10 West with some notable regional exceptions, like the Apollo Belenos from Aquileia and the Apollo Grannus from Grand: for epigraphic attestations cf. in general Jufer & Luginbühl 2001; also cf. the astrological tablets from Grand: Buisson & Abry 1993. Minerva is comparatively rare, suggesting that 11 Caesar’s interpretatio Romana did perhaps not usually relate to how many local people saw their own deities. [5] seiliedig ar ei wybodaeth bersonol am Belenos a Grannus y Celtiaid?10 A phwy yw Minerva Cesar yng Ngwlad Gâl, o ystyried bod mwy o dystiolaeth ynghylch duwiesau megis Iwno, Maia/ Rosmerta a’r amryfal ‘famau’?11 Mae hyn yn codi inter alia ddau gwestiwn pwysig: Mae interpretationes Cesar yn datgelu problem trosi; a ellir cael dau dduw hollol gyfatebol mewn dau wahanol ddiwylliant neu ddwy wahanol grefydd? Na ellir. O’r herwydd, byddai gwahanol bobl wedi gwneud cysylltiadau gwahanol yn ôl eu gwybodaeth bersonol rhwng y duwiau Celtaidd, Groegaidd a Rhufeinig yn ôl eu gwybodaeth bersonol am yr addoliad neu’r duw lleol, ar eu dehongliad personol, ac yn dibynnu ar amser, gan fod ystyr addoliad yn rhwym o newid dros amser. Pa mor ddibynadwy yw adroddiadau cyffredinoli Cesar? Gwnaed interpretatio pobl leol yn newydd: CIL XII 3090 = 4136 = yn enwedig I 441. Ond mae modd i ni gysylltu’r mathau hyn o dduwiau Mercher ‘Rhufeinig-Geltaidd’ ag unrhyw dduw Celtaidd cyn-Rufeinig (neu ai fersiwn fwy heddychol addas ar gyfer y Dywysogaeth Rufeinig sydd gennym?). Mae Scholies Berne ar Pharsalia Lucan (a hefyd Mawrth!) yn cysylltu Esus a Toutatis (Zwicker 1935, I, 51–2; cf. Duval 1989 am drafodaeth o ffynonellau hynafol ar gyfer Esus, Toutatis a Taranis). Ond mae hyn yn codi’r cwestiwn unwaith yn rhagor a yw’r geiriau Celtaidd hyn yn cyfeirio at dduw, neu a ydynt yn cael eu defnyddio yn gyfenw neu’n deitl gydag esus yn golygu ‘duw’ yn unig (cf. Meid 2003) a toutatis yn golygu ‘o blith y bobl’ (Delamarre 2003, 294–5). Mewn gwirionedd, mae Apolon braidd yn brin ar draws y Gorllewin Rhufeinig gyda rhai eithriadau rhanbarthol lleol megis yr Apolon Belenos o Acwileia a’r Apolon Grannus o Grand: i gael tystiolaeth epigraf ig cf. yn gyffredinol Jufer a Luginbühl 2001; hefyd cf. y llechi seryddol gan Grand: Buisson ac Abry 1993. Mae Minerva yn gymharol brin; efallai nad oedd interpretatio Romana Cesar yn adrodd yn aml y modd oedd pobl leol yn gweld eu duwiau eu hunain. [6] 12 13 14 I. c e lt ic r e l igions in th e Rom a n pe r io d the important god from which all Gauls are said to be descendent, was identi ied as Dispater by Caesar (BGall. 6.18); the evidence suggests that this was the mallet god, well known from iconography; a few worshippers provide us with a Celtic name, Sucellos, ‘the good striker’,12 but he is frequently referred to on inscriptions as Silvanus, not Dispater. Local people obviously made different associations and identi ications between their own gods and the Roman ones. As a result some deities are rather popular in the Roman West, like Hercules (cf. Paola Tomasi’s paper on Hercules in Cisalpine Gaul).13 Some people might even make more sophisticated interpretationes, displaying their intrinsic knowledge of Graeco-Roman culture and mythology; for instance, what is a dedication to Latona, the Greek Letô, doing at Montmirat in Nîmes/Nemausus’ rather remote hilly hinterland? It probably was an individual’s educated explanation for the local mother goddess that went beyond the standard mater or Minerva.14 Another dedicant called Of course, sucellos, ‘the good striker’, might only have been used as an epithet or nickname to describe the major attribute of this impor- 12 tant god: the mallet (which in turn might have prompted Caesar’s interpretatio, comparing him to the Etruscan Charu known from gladiatorial games: cf. Haeussler 2012, 152–4). Also cf. the Irish Donn, ‘the dark one’, and Welsh Beli Mawr, ‘Beli the Great’, which in turn might relate to the god Belenos. For Hercules in Gaul, cf. e.g. Bauchhenß 2008 and Haeussler 2012. For the distribution of 13 Hercules dedications in Cisalpine Gaul, also cf. Haeussler 2015. AE 2007, 929; single mother goddesses are quite common in this region, but often referred 14 to as terra mater (e.g. at Clarensac: CIL XII 4140) or Minerva (at Combas: ILGN 385–6). yr Ymerodraeth Rufeinig, gan wyro yn amlwg o ddehongliad Cesar: er enghraifft, nododd Cesar mai’r duw pwysig y dywedid bod pawb yng Ngwlad Gâl yn ddisgynnydd iddo oedd Dispater (BGall. 6.18); awgryma’r dystiolaeth mai y duw-ordd oedd hwn, sy’n enwog iawn mewn eiconograf ig: gan rai addolwyr ceir enw Celtaidd arno, sef Sucellos, ‘yr ergydiwr da’, 12 ond ar arysgrifau cyfeirir ato yn aml yn Silvanus, yn hytrach na Dispater. Yn amlwg, gwahanol oedd y cysylltiadau a’r uniaethu a wnâi pobl leol rhwng eu duwiau hwy a rhai’r Rhufeiniaid. O ganlyniad, mae rhai duwiau’n eithaf poblogaidd yng ngorllewin yr Ymerodraeth, megis Ercwlff (cf. papur Paola Tomasi ar Ercwlff yng Ngâl Isalpaidd).13 Gallai rhai pobl hyd yn oed wneud interpretationes mwy sof istigedig, gan arddangos eu gwybodaeth gynhenid am ddiwylliant Groeg a Rhufain a mytholeg; er enghraifft, beth mae cysegriad i Latona, sef Letô’r Groegiaid, yn ei wneud ym Montmirat yng nghefnwlad eithaf mynyddig Nîmes/Nemausus’? Mae’n debyg mai cynnig addysgedig unigolyn ar gyfer y fam dduwies leol a aeth y tu hwnt i’r mater neu’r Minerva safonol.14 Wrth gwrs, efallai mai dim ond teitl neu lysenw oedd sucellos, ‘yr ergydiwr da’, i ddisgri io un o briodoleddau mawr y duw pwysig hwn: yr ordd (a allai yn ei dro wedi sbarduno interpretatio Cesar, a’i cymharodd â charu’r Etrwsgiaid a oedd yn gyfarwydd o gemau’r cleddyfwyr: cf. Haeussler 2012, 152–4). Hefyd cf. Donn y Gwyddelod, ‘yr un du’, a Beli Mawr y Cymry, a allai yn ei dro ymwneud â’r duw Belenos. Ar gyfer Ercwlff yng Gwlad Gâl, cf. e.e. Bauchhenß 2008 a Haeussler 2012. Ar gyfer dosbarthiad cysegriadau Ercwlff yng Ngwlad Gâl Isalpaidd, hefyd cf. Haeussler 2015. AE 2007, 929; mae duwiesau mam sengl yn eithaf cyffredin yn y rhanbarth hwn, y cyfeirir ato yn aml yn terra mater (e.e. yn Clarensac: Hae ussle r & King 15 the local goddess at Lioux (Vaucluse) [---]ronea,15 probably identifying the local goddess with the ItaloEtruscan fertility goddess Feronea. In addition, across the empire our epigraphic record must be the result of countless religious and cultural misunderstandings: how many people – locals, pilgrims, migrants of various origins – might have misunderstood the meaning of a cult, of a theonym, and how to pronounce it? And how many people (and stone masons) were trying to unconsciously ‘Latinise’ unfamiliar theonyms? Still in today’s polytheistic religions, many people travel far to speci ic shrines, for example to participate in particular festivals or to pray for a good new year, but their knowledge of a deity and his/her myth is often limited to the particular function they are interested in – after all, there are no dogmas, no canonical sacred texts. How much more must this have been the case in Antiquity? At different periods of the year, people visited a sanctuary for particular reasons, and this can result in diverging, or even con licting, evidence. Also, we have to ask which deities are attested more frequently, and what this signi ies. Taking into account that our available epigraphic testimonies primarily consist of personal exvotos, we can expect that the deities mentioned re lect people’s personal ILN-4, 136: []roneai. Feronea is the only known theonym that would make sense here. The other inscriptions from Lioux refer to M() or M[---], usually interpreted as Mars (ILN-4, 135, 139–40), but other deities, like Mercury and Minerva, are also feasible. [7] Galwodd cysegrwr arall y dduwies leol yn Lioux (Vaucluse) yn [---] ronea,15 gan uniaethu’r dduwies leol â’r dduwies ffrwythlonrwydd ItaligEtrwsgaidd Feronea. Yn ogystal, ar draws yr ymerodraeth rhaid i’n cofnod epigraf ig fod o ganlyniad i gamgymeriadau crefyddol a diwylliannol dirifedi: faint o bobl – pererinion lleol, ymfudwyr o wahanol darddiadau – a allai wedi camddeall ystyr addoliad, enw duw, a sut i’w ynganu? A sut oedd rhai pobl (a seiri cerrig) yn ceisio ‘Lladineiddio’ enwau duwiau anghyfarwydd? Hyd yn oed heddiw mewn crefyddau polytheistaidd, mae llawer o bobl yn teithio i allorau penodol, er enghraifft cymryd rhan mewn gwyliau penodol neu weddïo am lwyddyn newydd dda, ond yn aml mae eu gwybodaeth am dduw/ies a’i f/myth yn aml yn gyfyngedig i’r swyddogaeth benodol y mae ganddynt ddiddordeb ynddi – wedi’r cyfan, nid oes unrhyw ddogmâu, nac unrhyw destunau sanctaidd canonaidd. Pa faint mwy gwir fyddai hyn wedi bod yn yr Hen Fyd? Ar wahanol adegau o’r lwyddyn, byddai pobl yn ymweld â chysegr am resymau penodol, a gall hyn arwain at dystiolaeth sy’n dargyfeirio neu hyd yn oed sy’n gwrthdaro. Hefyd, rhaid i ni ofyn pa dduwiau y ceir tystiolaeth iddynt fynychaf, ac arwyddocâd hyn. Gan 15 CIL XII 4140) neu Minerva (yn Combas: ILGN 385–6). ILN-4, 136: []roneai. Feronea yw’r unig enw duw hysbys a fyddai’n gwneud synnwyr yma. Mae’r arysgrifau eraill o Lioux yn cyfeirio at M() neu M[---], a ddehonglid fel arfer ym Mawrth (ILN-4, 135, 139–40), ond mae duwiau eraill, megis Mercher a Minerva, hefyd yn bosibl. [8] I. c e lt ic r e l igions in th e Rom a n pe r io d needs. We should therefore expect an unrepresentative preponderance of deities relating to healing, fertility, and prosperity in our record. In addition there might also be many cults associated with natural features, like springs and mountains, where Latin speakers preserved (a version of) the indigenous name since it could not be translated. ‘Pan-Celtic’ deities, and those related to myth, might feature less in our inscriptions as these might have been above all important for collective events and rarely resulted in personal ex-votos; Vienne provides us with an exceptional case where civic cults played a rather prominent role and many dedications were set up by magistrates and priests of the colonia:16 accordingly the selection of recorded deities differ, with Mercury and Apollo playing a more conspicuous role, while there is even a unique municipal priesthood for Mars, the lamen Martis, showing the importance of his cult for the Allobrogi and entire territory of the colonia Vienna, probably going back to pre-Roman times, perhaps as protector of the ‘tribe’.17 16 17 On the difference in the epigraphic habit of Vienne and other ciuitates and coloniae in Gal- 16 lia Narbonensis, cf. Haeussler 2014b: the evidence from Vienne and her territory seems to have been much more the result of municipal magistrates and priests making dedications on behalf of the community, vicus, or colonia. Also, many deities were given the name augustus, not just ‘Roman’ gods, but also e.g. the Matres Augustae, the ‘august mothers’: ILN-5.1, 13–16. The designation toutatis, meaning ‘of the tribe/ 17 state’, might be appropriate here. gymryd i ystyriaeth fod y tystiolaethau epigraf ig sydd ar gael gennym yn cynnwys offrymau personol yn bennaf, gellir disgwyl bod y duwiau a enwir yn adlewyrchu anghenion personol pobl. Dylem, felly, ddisgwyl mwyafrif anghynrychiadol o dduwiau sy’n ymwneud ag iacháu, ffrwythlonrwydd a llewyrch yn ein cofnod. Yn ogystal, gall hefyd fod llawer o ddefodau sy’n gysylltiedig â nodweddion lleol, megis ffynhonnau a mynyddoedd, lle cadwodd siaradwyr Lladin (fersiwn) o’r enw brodorol gan nad oedd modd cy ieithu’r enw brodorol. Gallai duwiau ‘PanGeltaidd’, a’r rheini sy’n gysylltiedig â myth, fod yn llai amlwg yn ein harysgrifau gan ei bod yn bosibl i’r rhain fod yn bwysig yn anad dim ar gyfer digwyddiadau cyffredinol heb arwain yn aml at offrymau personol; yn Vienne ceir achos eithriadol lle chwaraeodd defodau dinesig ran braidd yn amlwg a sefydlwyd nifer o gy lwyniadau gan ynadon ac offeiriaid y colonia:16 o’r herwydd, mae’r detholiad o dduwiau a gofnodir yn amrywio, gyda Mercher ac Apolon yn chwarae rhan amlycach, a hyd yn oed offeiriadaeth ddinesig unigryw ar gyfer Mawrth, y lamen Martis, sy’n dangos pwysigrwydd y ddefod hon i’r Allobrogi a holl diriogaeth y colonia Vienna, sy’n mynd yn ôl i’r adeg cyn y Rhufeiniaid fwy na thebyg, ac efallai iddo fod yn un o amddiffynwyr y ‘llwyth’.17 Ar y gwahaniaeth rhwng arfer epigraf ig Vienne a ciuitates a coloniae eraill yn Gallia Narbonensis, cf. Haeussler 2014b: ymddengys bod y dystiolaeth o Vienne a’i thystiolaeth yn fwy o lawer oherwydd bod yr ynadon a’r offeiriaid trefol yn gwneud cysegriadau ar ran y gymuned, vicus, neu’r colonia. Hefyd, cafodd nifer o dduwiau yr enw augustus, nid yn unig dduwiau ‘Rhufeinig’, ond hefyd e.e. Matres Augustae, ‘y mamau awstinaidd’: ILN-5.1, 13–16. Efallai y byddai’r cysegriad o’r enw toutatis, sef Hae ussle r & King 18 19 20 [ 9 ] Beyond the epigraphic record Y tu hwnt i’r cofnod epigraf g We can counterbalance some of the epigraphic shortcomings by studying iconography, sculpture, and art. Not only was iconography used widely, for example on anepigraphic altars and in rural areas, but iconography might also reveal important and powerful (civic) cults that left hardly any trace in the epigraphic record. Iconography, sculpture, and bas-reliefs can be used in public sanctuaries to present an anthropomorphic representation for a deity or to narrate a myth to the general public. In this respect, we can only speculate on the meaning of the widespread Jupiter giant columns that can be found in public places and at the centre of civic cults,18 or the representation of Mars and a bull as the local deities from Allones.19 We also need to ask whether we can still ind any traces of the so-called ‘druidic religion’ which, having been demonized between Tiberius and Claudius, used to be rather admired by Romans in the late Republic, including Caesar: druids as augurs, philosophers, and ‘Pythagoreans’.20 We might Gellir goresgyn rhai o’r diffygion epigraf ig trwy astudio eiconograffeg, cer lunio, a cher lunio. Yn ogystal â chael ei defnyddio’n eang, er enghraifft ar allorau anepigraf ig ac mewn ardaloedd gwledig, ond gallai eiconograffeg hefyd ddatgelu defodau (dinesig) pwysig a phwerus na adawsant fawr ddim ôl yn y cofnod epigraf ig. Gellir defnyddio eiconograffeg, cerfluniau, a cherfweddau isel mewn cysegrau i gy lwyno cynrychiolaeth anthropomorffaidd ar gyfer duw neu er mwyn adrodd chwedl i’r cyhoedd. Yn hyn o beth, rhaid dyfalu yn unig beth oedd ystyr y colofnau cawraidd Iau eang y gellir dod o hyd iddynt mewn lleoedd cyhoeddus ac ynghanol defodau dinesig,18 neu gynrychioliadau o Fawrth ar ffurf tarw yn yr un modd â’r duwiau lleol o Allones.19 Mae hefyd angen i ni holi sut y gallwn ni ddarganfod unrhyw olion o ‘grefydd y derwyddon’ a gollfarnwyd rhwng Tiberius a Claudius, ond a edmygid gan Rufeiniaid yn y Weriniaeth hwyr, gan gynnwys Cesar: derwyddon yn ddaroganwyr, athronwyr a ‘Pythagoreaid’.20 Yma gellir cyfeirio at Cf. for example the number of Jupiter columns that were identi ied on the cathedral mount at Worms/Borbetomagos, suggesting a civic 18 cult of the civitas, and a similar concentration in Alzey: cf. Haeussler 2008c. Also see Florian Blanchard’s paper in this volume. Cf. interpretation of Gury 2012 on the sculpture from the Mars Mullo cult at Allonnes. Cf. Cicero, De divinatione I 41 [90] (44 B.C.), 19 on the Aeduan druid Divitiacus, comparing him to Roman augurs and Persian magi: ‘He 20 (Divitiacus) claimed to have that knowledge of nature which the Greeks call “physiologia”, and he used to make predictions, sometimes by means of augury and sometimes by means of conjecture’. Also cf. Strabo 4.4; Caesar BGall. ‘gan y llwyth/y wladwriaeth’, yn briodol yma. Cf. er enghraifft nifer y colofnau Iau a nodwyd ar fryn y gadeirlan yn Worms/Borbetomagos gan awgrymu addoliad dinesig y civitas, a ffocws tebyg yn Alzey: cf. Haeussler 2008c. Hefyd gweler papur Florian Blanchard yn y gyfrol hon. Cf. dehongliad Gury 2012 ar y cer lun o addoliad Mars Mullo yn Allones. Cf. Cicero, De divinatione I 41 [90] (44 B.C.), ar dderwydd Aeduan Divitiacus, gan gymharu daroganwyr Rhufain a doethion Persia: ‘Honnodd (Divitiacus) fod ganddo wybodaeth am natur sef yr hyn a eilw’r Groegiaid yn “ffysiologia”, ac arferai ragddweud, weithio trwy ddaro- [ 10 ] I. c e lt ic r e l igions in th e Rom a n pe r io d allude here to the famous calendar of Coligny of the late 2nd century AD, a major epigraphic source for our understanding of ‘Celtic’ religion and major seasonal events.21 This leads us to the role of astronomy: the paper by Burillo-Mozota and BurilloCuadrado on the latest excavations of the Celtiberian sanctuary at Segeda reveals once more the importance of cosmological aspects in pre-Roman cults. We can identify such aspects, in various forms, also in other regions of the Keltiké, for example in the popularity of gods, such as Sol and Luna (though rarely with Celtic name), as well as in the astronomical orientation of temples, sanctuaries, houses, and tombs throughout the Roman period.22 To understand ‘Celtic’ cults therefore goes beyond the names of deities. Bone evidence, too, can reveal a cult’s religious calendar with animal sacri ice taking place at particular points in the calendar; certain preRoman and Roman-period deities, for example, might require sacri ices at samonios/Samhain, for example at Hayling Island and Acy-Romance, others at Bealtaine or Lughnasa.23 It is remarkable to see the persistence of such rituals into the Roman period, even though a cult might have acquired a more ‘Roman’ appearance in time. 21 22 23 6.14; Amm. Marc. 15.9.8; for an overview, cf. Haeussler 2014, 48. For a critical edition of the calendar of Coligny, 21 see RIG-3; also cf. Swift 2002. Cf. C. Haselgrove 1995 with further biblio- 22 graphy. Cf. Green 1992, 42, s.v. Beltene; 136, s.v. Lugh- 23 nasad; 185–6, s.v. Samhain. For Hayling Island, cf. King 2005, 363, for Acy-Romance, cf. Lambot 2006, 185. galendr enwog Coligny ar ddiwedd yr 2il ganrif OC, ffynhonnell epigraf ig bwysig ar gyfer ein dealltwriaeth o’r grefydd ‘Geltaidd’ a digwyddiadau tymhorol o bwys.21 Mae hyn yn ein harwain ni at rôl seryddiaeth: mae’r papur gan Burillo-Mozota a BurilloCuadrado ar y cloddiadau diweddaraf yn y gysegrfa Geltiberaidd yn Segeda yn egluro unwaith eto bwysigrwydd agweddau cosmolegol ar ddefodau cyn-Rufeinig. Gallwn nodi’r agweddau hyn, ar wahanol ffur iau, hefyd mewn rhanbarthau eraill o’r Keltiké, er enghraifft poblogrwydd y duwiau, megis Sol a Luna (er yn anaml gydag enw Celtaidd), yn ogystal â gogwydd seryddol temlau, cysegrau, tai a beddau gydol y cyfnod Rhufeinig.22 Mae deall defodau ‘Celtaidd’, felly, yn mynd y tu hwnt i enwau’r duwiau. Gall tystiolaeth esgyrn, hefyd, ddatgelu calendr defod gydag aberthau anifeiliaid yn digwydd ar adegau penodol yn y calendr; gallai fod gofyn i rai o dduwiau’r cyfnod cyn-Rufeinig a Rhufeinig gael aberthau yn samonios/ Samhain, er enghraifft ar Ynys Hayling ac Acy-Romance, ac eraill yn Bealtaine neu Lughnasa. 23 Mae’n rhyfedd gweld parhad y defodau hyn yn y cyfnod Rhufeinig, hyd yn oed pe bai defod wedi magu gwedd fwy ‘Rhufeinig’ gydag amser. Yn absenoldeb arysgrifau, y gan ac weithiau trwy ddyfalu’. Hefyd cf. Strabo 4.4; Cesar BGall. 6.14; Amm. Marc. 15.9.8; i gael trosolwg, cf. Haeussler 2014, 48. Am argraf iad beirniadol o galendr Coligny, gweler RIG-3; hefyd cf. Swift 2002. Cf. C. Haselgrove 1995 gyda llyfryddiaeth bellach. Cf. Green 1992, 42, s.v. Beltene; 136, s.v. Lughnasad; 185–6, s.v. Samhain. Ar gyfer Ynys Hayling Island, cf. King 2005, 363, ar gyfer AcyRomance, cf. Lambot 2006, 185. Hae ussle r & King 24 In the absence of inscriptions we can often only speculate about the deity (or deities) that received particular sacri ices at particular times of the year, but it would be interesting to see if patterns can be identi ied across the Keltiké; after all, a certain standardisation of cult architecture in the Roman period (the so-called Romano-Celtic temple, Umgangstempel or fanum, relatively square temples, usually with an ambulatory, quite unlike a Roman temple) shows an astonishing degree of religious communication across the provinces. And we may wonder whether we are merely dealing here with an architectural ‘fashion’ or whether this was the result of the continued perseverance of a priestly ‘caste’?).24 With respect to the druids, the graf ito DRU on a 2nd-century AD incense burner from Chartres is equally intriguing as it may have referred to druids – and if so, we have to ask what such a ind really testi ies: the persistence (or ‘revival’?) of druids and druidic religion (cf. paper by Miranda Aldhouse-Green)? Can we really assume that all traces of ‘druidism’ were eradicated in the 1st century AD? Many pre-Roman priesthoods might have been transformed into Romanstyle priesthoods, like the lamen Martis of the Allobrogi (v. supra); in other cases, personal names might re lect people’s religious role, like the cognomen bardus, ‘bard(?)’, as in Lucius 24 For ‘Romano-Celtic temples’, cf. King 2007 for the origin of this architectural form; for a survey of all potential Romano-Celtic temple sites, 25 cf. Fauduet 2010. [ 11 ] cyfan y gellir ei wneud yw dyfalu am y duw (neu’r duwiau) a gafodd aberthau penodol ar adegau penodol yn ystod y lwyddyn, ond byddai’n ddiddorol gweld a ellir nodi patrymau ar draws y Keltiké; wedi’r cyfan, mae rhywfaint o safoni o ran pensaernïaeth y cyfnod Rhufeinig (sef y –deml RhufeinigGeltaidd, Umgangstempel fel y’i gelwir neu’r fanum, temlau cymharol sgwâr, fel arfer gyda rhodfa, yn wahanol iawn i deml Rufeinig) yn dangos gradd syfrdanol o gyfathrebu crefyddol ar draws y taleithiau. Ac efallai y gellir holi ai dim ond ‘ffasiwn’ pensaernïol a geir yma neu a oedd hyn o ganlyniad i barhad ‘cast’ offeiriadol?).24 O ran y derwyddon, mae’r graf ito DRU ar losgydd thus o’r 2il ganrif OC o Chartres yr un mor ddiddorol gan ei fod wedi cyfeirio’n wreiddiol at dderwyddon, o bosibl – ac os felly, rhaid gofyn beth a ddangosir gan y canfyddiad hwn: parhad (neu ‘adfywio’?) derwyddon a chrefydd y derwyddon (cf. papur gan Miranda Aldhouse-Green)? A ellir rhagdybio’n wir y dilëwyd holl olion ‘derwyddiaeth’ yn y ganrif 1af OC? Mae’n bosibl i nifer o fathau cyn-Rufeinig o offeiriadaeth gael eu trawsnewid yn rhai ar ddull Rhufeinig, megis lamen Martis yr Allobrogi (v. supra): mewn achosion eraill, gallai enwau personol adlewyrchu rôl grefyddol pobl, megis y cognomen bardus, ‘bardd(?)’, megis yn Lucius Erax Bardus a wnaeth gysegriad i Apolon Belenos yn Bardonecchia.25 Fel Ar gyfer ‘temlau Rhufeinig-Geltaidd’, cf. King 2007 ar gyfer tarddiad y ffurf bensaernïol hon: ar gyfer arolwg o holl ddarpar sa leoedd temlau Rhufeinig-Geltaidd, cf. Fauduet 2010. Yn yr Alpes Cottiae; cf. AE 1959, 170 = AE 2005, [ 12 ] 25 26 27 I. c e lt ic r e l igions in th e Rom a n pe r io d Erax Bardus who made a dedication to Apollo Belenos at Bardonecchia.25 Otherwise, we might look at traces of ‘magic’ in our search for survival of Iron Age rituals in the Roman period – but then we have to wonder whether ‘druidism’ had anything to do with magic, and if so, how can we distinguish Celtic, Roman, Greek, Egyptian magical traditions within an increasingly cosmopolitan Roman empire? arall gallem ni nodi rhai olion o ‘hud’ wrth chwilio am oroesiad defodau Oes Haearn yn y cyfnod Rhufeinig – ond wedyn rhaid i ni holi a oedd a wnelo ‘derwyddiaeth’ unrhyw beth â dewiniaeth, ac os felly, sut y gellir gwahaniaethu traddodiadau Celtaidd, Rhufeinig, Groegaidd, Eifftaidd mewn ymerodraeth Rufeinig a oedd yn fwy ac yn fwy cosmopolitan? Celticity & Translatability Celtigrwydd a Throsiadwyedd From its outset, the scope of F.E.R.C.AN. has always been much wider than just inscriptions for an obvious reason. Though there is a wealth of epigraphic evidence, we can only aspire to understand it if we take into account the wider context, like iconography and our archaeological knowledge of cult places, ritual practices, and sacred landscapes. Since its launch in 1998, ifteen workshops, organised locally by the F.E.R.C.AN. coordinators of the project’s European members (see Table 1), have provided an important venue to exchange ideas and enhance our methodology regarding ‘Celtic’ religions during the Iron Age and the Roman period. Needless to say that the de initions of ‘Celtic’ and ‘Celticity’ have always been a challenging issue.26 On the one hand, hundreds of deities have names in a Celtic language.27 It is O’r cychwyn cyntaf, mae cwmpas F.E.R.C.AN. wedi bod erioed yn ehangach o lawer nag arysgrifau yn unig, a hynny am resymau amlwg. Er bod cyfoeth o dystiolaeth epigraf ig, yr unig obaith sydd gennym o’i deall yw trwy gymryd i ystyriaeth y cyddestun ehangach, yn yr un modd ag eiconograf ig a’n gwybodaeth archeolegol am fannau addoli, arferion defodol a thirweddau cysegredig. Ers ei lansio yn 1998, mae tri ar ddeg o weithdai, a drefnwyd yn lleol gan gydlynwyr F.E.R.C.AN. ar gyfer aelodau Ewropeaidd y prosiect (gweler Tabl 1), wedi darparu lleoliad pwysig i wella ein methodoleg ynghylch crefyddau ‘Celtaidd’ yn ystod yr Oes Haearn a’r cyfnod Rhufeinig. Afraid dweud mai pwnc heriol fu dif inio ‘Celtaidd’ a ‘Celtigrwydd’ erioed.26 Ar y naill law, mae gan gannoedd o dduwiau enwau mewn iaith Geltaidd.27 Mae’n debyg y gallai rhai o’r rhain fod yn gysylltiedig In the Alpes Cottiae; cf. AE 1959, 170 = AE 2005, 961. Also cf. e.g. Q(uintus) Cassius Bardus (ILAfr 166, 8). Cf. Sims-Williams 1998 on Celtomania and Cel- 26 toscepticsm. Cf. in general the repertoire of Celtic deities in 27 Jufer and Luginbühl 2001. 961. Hefyd cf. e.e. Q(uintus) Cassius Bardus (ILAfr 166, 8). Cf. Sims-Williams 1998 ar Geltomania a Cheltosgeptiaeth. Cf. e.e. cyfres y duwiau Celtaidd yn Jufer a Luginbühl 2001. Hae ussle r & King feasible that some of them might have related to religious understandings and myths that might have their roots in pre-Roman times. But most of all, they illustrate that people still spoke Celtic dialects across the Roman West, enabling them to use (and hypothetically also create) theonyms and epithets in a Celtic language.28 In other words, Celtic theonyms may not necessarily indicate a pre-Roman ‘Celtic’ deity, especially when they just have rather literal meanings, like ‘mighty’ or ‘powerful’ (mogons / mogetios, vernostonos, etc.),29 and can therefore be associated as unspeci ic epithets to different gods.30 We have to be aware that the story is extremely complex: Celtic terms were used to describe deities in various forms, not just as theonyms, but also as epithets, epicleses, and cognomina; some names might make a reference to a deity’s particular function or characteristic, or to a mythical event. For instance, (Mars) Nabelcus, ‘The Wounder of Heaven’ or (Mars) Divannos, ‘The Great Destroyer’ might refer to creation myths.31 Similarly, 28 29 30 31 Cf. introduction to the 2005 F.E.R.C.AN. workshop in London: Haeussler and King 2007. For the problem of terminology, see De Bernardo Stempel 2008b . Cf. example mogons which appears on its own as singular and plural, male and female, since it is merely a reference to the ‘mighty’ god(s)/ goddess(es); there is also a dedication to the ‘mighty god V.’ (deo mogont(i) Vitire(!), from Netherby, RIB 971) and to the ‘mighty god C.’ ([d]eo mogonito(!) Cad(), from Risingham/ Habitancum, RIB 1225). For the term ‘unspecific epithet’, cf. e.g. De Bernardo Stempel 2008b, 69. Compare for example Shiva in Hinduism as the destroyer and creator of the world. 28 29 30 31 [ 13 ] â dealltwriaeth grefyddol ac efallai y gallai gwreiddiau rhai o’r chwedlau fynd yn ôl i amserau cyn-Rufeinig. Ond yn bennaf, dangosant fod pobl yn dal i siarad tafodieithoedd Celtaidd ar draws y Gorllewin Rhufeinig, gan eu galluogi nhw i ddefnyddio (ac mewn theori creu hefyd) enwau duwiau theitlau mewn iaith Geltaidd.28 Mewn geiriau eraill, efallai nad yw enwau duwiau, o reidrwydd, yn arwydd o dduwdod ‘Celtaidd’ cyn-Rufeinig, yn enwedig os ystyron braidd yn llythrennol sydd iddynt, megis ‘grymus’ neu ‘pwerus’ (mogons / mogetios, vernostonos, ac ati),29 ac o’r herwydd gellir eu cysylltu’n deitlau amhenodol ar gyfer gwahanol dduwiau.30 Rhaid i ni fod yn ymwybodol bod yr hanes yn un hynod gymhleth: defnyddid termau Celtaidd i ddisgri io duwiau ar amrywiol ffur iau, nid yn unig enwau duwiau, ond hefyd teitlau, arddeisy iadau, a llysenwau; gallai rhai enwau gyfeirio at swyddogaeth neu nodwedd benodol rhyw dduw neu’i gilydd, neu at ddigwyddiad mytholegol. Er enghraifft, gallai (Mawrth) Nabelcus, ‘Clwyfwr y Nefoedd’ neu (Mawrth) Divannos, ‘Y Dinistriwr Mawr’ gyfeirio at chwedlau creu.31 Yn yr un modd, Cf. cy lwyniad i weithdy 2005 F.E.R.C.AN. yn Llundain: Haeussler a King 2007. Ar gyfer y broblem, erminoleg De Bernardo Stempel 2008b . Cf. yr enghraifft mogons sy’n ymddangos ar ei phen ei hun ar ffurf unigol neu luosog, yn wrywaidd ac yn fenywaidd, gan mai cyfeiriad yn unig yw at y duw(iau)/duwies(au) ‘nerthol’; mae hefyd gyfeiriad at y ‘duw nerthol V.’ (deo mogont(i) Vitire(!), o Netherby, RIB 971) ac at ‘y duw nerthol C.’ ([d]eo mogonito(!) Cad(), o Risingham / Habitancum, RIB 1225). Ar gyfer y term ‘teitl amhenodol’, cf. e.e. De Bernardo Stempel 2008b, 69. Cymharer, er enghraifft, Shiva mewn Hindŵaeth yn ddinistrydd a chreawdwr y byd. [ 14 ] 32 33 I. c e lt ic r e l igions in th e Rom a n pe r io d names like Uxovinus, ‘The Very White’,32 might just refer to a white deity – or his sculpture – but could equally refer to a particular myth, perhaps re lecting the god’s purity and/or supreme knowledge. We constantly need to re-think our interpretations, and the larger our data base, the better our methodology might become. Moreover, we need to bear in mind that a clear attribution is not always possible: some ‘Celtic’ theonyms have equally been interpreted as Germanic, Aquitanian, or pre-Celtic.33 Some reason for this confusion is of course our source material: we primarily know these names from a Latin-speaking context, and usually only in the dative; hence also the dif iculty whether to reconstruct a Latin nominative in -us or a Celtic nominative in -os: should we 32 ILN-4, 124 = ILS 4694 from Apt; translation suggested by De Bernardo Stempel 2007. 33 There are different interpretations, e.g. for the Almahae (Celtic, Germanic, or Gallo-Latin, cf. Haeussler 2008 with further bibliography), or the deities in Aquitania (Celtic for De Bernardo Stempel 2008, Aquitanian for Gorrochateguí 2008; 2013). Finally, there is also the question of deonomastic names, i.e. theonyms or epithets named after a locality, town, river, mountain, ethnos, person, etc., sometimes preserving pre-Celtic toponyms in the theonym. But the question is of course what came irst: e.g., is the god Nemausus named after the city Nemausus (Nîmes) or vice versa? The god Vintur in Gallia Narbonensis, for example, seems to be the personi ication of the Mont Ventoux: consequently we might label Vintur a ‘deonomastic’ theonym, but this does not deny that people in the Roman period (and surely also in the late Iron Age) were worshipping and appeasing the god Vintur (similar to other mountainous deities, like Poeninus), i.e. the dedications in his honour are evidence for ‘Celtic’ (i.e. indigenous, local) religion (ILN-Apt, 17 (Apt), ILN-Apt, 143 (Goult), CAG-26, p. 421 (Mirabelaux-Baronnies)). mae’n bosibl bod enwau megis Uxovinus, ‘Y tra gwyn’, 32 yn cyfeirio at ryw dduw gwyn – neu ger lun ohono – ond mae’r un mor bosibl ei fod yn cyfeirio at chwedl benodol, gan adlewyrchu, o bosibl, purdeb y duw a/ neu ei wybodaeth aruchel. Mae angen yn barhaus i ni ailfeddwl ein dehongliadau, a pho fwyaf y gronfa ddata sydd gennym, gorau oll fydd ein methodoleg. Hefyd, mae angen i ni go io nad yw priodoli clir bob amser yn bosibl: mae rhai enwau duwiau ‘Celtaidd’ wedi eu dehongli i’r un graddau yn rhai Germanig, Acwitanaidd neu gyn-Geltaidd.33 Un rheswm dros y dryswch, wrth gwrs, yw ein ffynonellau: yn bennaf, rydym yn adnabod yr enwau hyn o gyd-destun Lladin, ac fel arfer ar y ffurf dderbyniol yn unig; o ble daw hefyd yr anhawster a ddylid ail-greu ffurf enwol Ladin gydag ILN-4, 124 = ILS 4694 o Apt; y cy ieithiad wedi ei awgrymu gan De Bernardo Stempel 2007. Mae gwahanol ddehongliadau, e.e. ar gyfer yr Almahae (Celtaidd, Almaenig, neu GalaiddLadin, cf. Haeussler 2008 gyda llyfryddiaeth bellach), neu’r duwiau yn Acwitania (Celtaidd ar gyfer De Bernardo Stempel 2008, Acwitanaidd ar gyfer Gorrochateguí 2008; 2013). Yn olaf, mae hefyd gwestiwn enwau deonomastig, h.y. enwau duwiau neu deitlau wedi eu henwi ar ôl ardal leol, tref, afon, ethnos, person, ac ati, sydd weithiau’n cadw enwau lleoedd cyn-Geltaidd yn enw’r duw. Fodd bynnag, y cwestiwn, wrth gwrs, yw pa un a ddaeth yn gyntaf: e.e., a yw’r duw Nemausus wedi ei enwi ar ôl y ddinas Nemausus (Nîmes) neu’r gwrthwyneb? Ymddengys mai personoli’r Mont Ventoux a wna’r duw Vintur yn Gallia Narbonensis, er enghraifft: o’r herwydd, byddai modd labelu Vintur yn enw duw ‘deonomastig’, ond nid yw hyn yn golygu nad oedd pobl yn y cyfnod Rhufeinig (ac ar ddiwedd yr Oes Haearn hefyd, mae’n rhaid) yn addoli ac yn heddychu’r duw Vintur (yn debyg i dduwiau mynyddig eraill, mae Poeninus), h.y. mae’r cysegriad er anrhydedd iddo yn dystiolaeth o grefydd ‘Geltaidd’ (h.y. brodorol, lleol) (ILN-Apt, 17 (Apt), ILN-Apt, 143 (Goult), CAG26, t. 421 (Mirabel-aux-Baronnies)). Hae ussle r & King 34 say, for instance, Cocidius or Cocidios? Moreover, the paper by Alves Dias and Correia Santos reminds us that not every Celtic name, like Danceroi and Aro, was necessarily a theonym – instead, a critical contextual analysis is essential to re-think traditional assumptions. On the other hand, we can also recognise further testimonies for these ‘indigenous’, ‘Celtic’ religions, as manifested in the Roman period. Mother goddesses, like the matres, matronae, Iunones, proxsumae, or single mothers, like terra mater or regina, despite their Roman name, clearly relate to local religious understandings, not Greek or Roman ones.34 Sometimes this becomes more For Matres, Iunones, etc. cf. overview in Haeussler 2008b with further bibliography; cf. Delamarre 2013 for the case of the Matronae 34 Andrusteihae within a Dumezilian tripartite division of the world. There are also ‘pre-Roman’ attestations in Gallo-Greek, probably dating no earlier than the 1st century BC, e.g. the mothers from Glanum and Nimes (v. supra); approximately 500 years ealier is the Tartessian inscription from southern Portugal that contains the divine name ekurine; John Koch has identi ied this as Epona regina or ‘horse queen’ (Koch 2013, 43–4, J.4.1; for Epona, v. supra). For the evidence on Juno Regina and Regina, cf. Haeussler 2008b, 26–7: e.g., altars to Regina from Leamington (Glos.) (RIB 125 = CSIR-GB I.7, 94: Dea Regina – in the nominative – with rather crude representation of the goddess) and Lanchester/Longovicium (RIB 1084: Reginae votum Misio v.l.s.), and regina was also associated with other Celtic names, e.g. with Candida: dea regina Candida (at Osterburken: AE 1985, 685, 695; Candida is considered a calque of a Celtic theonym for Lejeune 1981); on a graf ito in Celtic from Lezoux, we also ind the term rigani, now associated with Rosmerta, who is widely attested on Roman inscriptions: e ieuri Rigani Rosmertiae – for a discussion whether this is a dedication to ‘Queen Rosmerta’ or to the ‘Purveying Queen’, see Lejeune 1981; for Rosmerta as ‘La Pourvoyeuse’ cf. De- [ 15 ] -us neu ffurf enwol Geltaidd gydag -os: a ddylid dweud, er enghraifft, Cocidius neu Cocidios? Yn ogystal, mae’r papur gan Alves Dias a Correia Santos ein hatgoffa ni nad yw pob enw Celtaidd, megis Danceroi ac Aro, o reidrwydd yn enw duw – yn hytrach, rhaid wrth ddadansoddi cyd-destunol beirniadol i ailfeddwl rhagdybiaethau traddodiadol. Ar ben popeth arall, gellir hefyd gydnabod tystiolaeth bellach ar gyfer y crefyddau ‘brodorol’, ‘Celtaidd’ hyn a welwyd yn y cyfnod Rhufeinig. Mae mam-dduwiesau, megis y matres, matronae, Iunones, proxsumae, neu famau sengl, megis terra mater neu regina, er gwaethaf eu henw Rhufeinig, yn cyfeirio’n glir at ddealltwriaeth grefyddol leol, yn hytrach na rhai Groegaidd neu Rufeinig.34 Weithiau Ar gyfer Matres, Iunones, ac ati cf. y trosolwg yn Haeussler 2008b gyda llyfryddiaeth bellach; cf. Delamarre 2013 ar gyfer achos Matronae Andrusteihae, o fewn rhaniad triphlyg y byd gan Dumézil. Mae hefyd dystiolaeth ‘gyn-Rufeinig’ mewn Galaidd-Roeg, nad yw’n debygol o fod yn gynharach na’r ganrif gyntaf C.C. e.e. y mamau o Glanum a Nîmes (v. supra); tua 500 mlynedd ynghynt yn yr arysgrifen Dartesaidd o dde Portiwgal sy’n cynnwys yr enw dwyfol ekuurine; mae John Koch wedi nodi mai Epona regina neu ‘farch-frenhines’ (Koch 2013, 43–4, J.4.1; ar gyfer Epona, v. supra). Ar gyfer y dystiolaeth ar Juno Regina a Regina, cf. Haeussler 2008b, 26–7: e.e., allorau i Regina o Leamington (Caerloyw) (RIB 125 = CSIR-GB I.7, 94: Dea Regina – yn yr enwol – gyda chynrychiolaeth braidd yn amrwd o’r dduwies) a Lanchester/ Longovicium (RIB 1084: Reginae votum Misio v.l.s.), ac roedd regina hefyd wedi ei gysylltu ag enwau Celtaidd eraill, e.e. gyda Candida: dea regina Candida (yn Osterburken: AE 1985, 685, 695; ystyrir Candida yn ddynwarediad ar enw duw Celtaidd gan Lejeune 1981); mewn grafito Celteg o Lezoux, cawn hefyd y term rigani, sydd bellach wedi ei gysylltu â Rosmerta, y mae tystiolaeth eang iddo ar arysgrifau Rhufeinig: e ieuri Rigani Rosmertiae – am drafodaeth ai cysegriad i’r ‘Frenhines Rosmerta’ neu i’r [ 16 ] 35 36 37 38 I. c e lt ic r e l igions in th e Rom a n pe r io d apparent when Celtic epithets are used, like the ‘well-leading mother goddesses’, the Matres Suleviae,35 or the Matronae Dervonnae (from Celtic dervos, ‘oak’).36 And in this ‘global’ world of the Roman empire, we can sometimes see how people re lect on the divine variations they encounter across the empire, like soldiers from Britain making dedications to the matres transmarinis,37 ‘the mothers from overseas’, or to Mat(ribus) Af(ris) Ita(lis) Ga(llis), ‘the African, Italic, and Gallic mothers’.38 Iconography also reveals other ‘indigenous/Celtic’ deities and believes across the Keltiké in Roman times, like the mallet god, the wheel god, the various tree and warrior gods, and of course the ‘Jupiter Giant Rider’ (Jupitergigantenreiter or cavalier à l’anguipède), particularly typical for lamarre 2003, 276, smer, smerto-. Usually, people only referred to them as Suleviae – the Matres was probably implied – but there are dedications to the Matres Suleviae, notably from Rome and Colchester: cf. Jufer & Luginbühl 2001, 64, s.v. Suleviae. This epithet was also used for single goddesses, like the Sulevia Edennica Minerva from Collias (Gallia Narbonensis): CIL XII 2974. Matribus Dervonnibus at Milan/Mediolanum (CIL V 5791) and from nearby Brescia/Brixia, the deity’s name was ‘Latinised’ to Fatae whilst preserving the Celtic epithet/cognomen: Fatis Dervonibus (CIL V 4208 = Inscr.It. 10.5, 813 = ILS 3762). For dervos, cf. Delamarre 2003, 140, s.v. dervos. Also cf. other ‘tree deities’, like the Matres Baginatiae in the Drôme (AE 2000, 884–886). RIB 919, 920, 1030, 1224, 1318, 1989 from Old Penrith, Newcastle, Risingham, Binchester, and Castlesteads. RIB 653 = ILS 4787 from York, set up by the legionary Marcus Minucius Audens. The ‘Italic mothers’, probably referring to Cisalpine Gaul, are also evoked in a dedication of an altar from Dover: RIB III 3031, set up by a st(rator) co(n)[s(ularis)] named Ol(us) Cor[---] Candid(us). daw hyn i’r amlwg wrth ddefnyddio teitlau Celtaidd, megis ‘y mamdduwiesau ffynnon-arweiniol’, y Matres Suleviae,35 neu’r Matronae Dervonnae (o’r gair Celtaidd dervos, ‘derwen’).36 Ac yng nghyd-destun ‘byd-eang’ yr Ymerodraeth Rufeinig, weithiau gellir gweld sut mae pobl yn adfyfyrio ar yr amrywiadau dwyfol y maent yn dod ar eu traws ledled yr ymerodraeth, megis milwyr o Brydain sy’n cysegru i’r matres transmarinis37 ‘y mamau o dramor’, neu i Mat(ribus) Af(ris) Ita(lis) Ga(llis), ‘y mamau Affricanaidd, Eidalaidd a Galaidd’.38 Trwy eiconograf i daw hefyd dduwiau a chredoau ‘brodorol/ Celtaidd’ eraill i’r amlwg ar draws y Keltiké adeg y Rhufeiniaid, megis y duw-ordd, y duw-olwyn, y gwahanol dduwiau-goed a duwiau-ryfelwyr, ac 35 36 37 38 ‘Frenhines sy’n Darparu’ sydd dan sylw, gweler Lejeune 1981; ar gyfer Rosmerta yn ‘La Pourvoyeuse’ cf. Delamarre 2003, 276, smer, smerto-. Fel arfer, ni fyddai pobl ond yn cyfeirio atynt yn Suleviae – mae’n debyg bod Matres ymhlyg – ond mae cysegriadau i’r Matres Suleviae, yn bennaf o Rufain a Colchester: cf. Jufer a Luginbühl 2001, 64, s.v. Suleviae. Rhoddwyd y teitl hwn hefyd i dduwiesau unigol, megis y Sulevia Idennica Minerva o Collias (Gallia Narbonensis): CIL XII 2974. Matribus Dervonnibus ym Milan/Mediolanum (CIL V 5791) ac o Brescia/Brixia gerllaw, ‘Lladineiddiwyd’ enw’r duw yn Fatae gan gadw’r teitl/cyfenw Celtaidd: Fatis Dervonibus (CIL V 4208 = Inscr.It. 10.5, 813 = ILS 3762). Ar gyfer dervos, cf. Delamarre 2003, 140, s.v. dervos. Hefyd cf. ‘coed-dduwiau’ eraill, megis y Matres Baginatiae yn y Drôme (AE 2000, 884–886). RIB 919, 920, 1030, 1224, 1318, 1989 o Old Penrith, Newcastle, Risingham, Binchester, a Castlesteads. RIB 653 = ILS 4787 o Gaerefrog, a sefydlwyd gan y lleng ilwr Marcus Minucius Audens. Gelwir hefyd ar y ‘mamau Italig’, gan gyfeirio fwy na thebyg at Gâl Isalpaidd, mewn cysegriad allor o Dover: RIB III 3031, a sefydlwyd gan st(rator) co(n)[s(ularis)] o’r enw Ol(us) Cor[---] Candid(us). Hae ussle r & King 39 eastern Gaul, whose relationship to Gallo-Roman religions is analyzed by Florian Blanchard in this volume. Though sometimes associated with theonyms (or epithets) in a Celtic language, like sucellos (‘the good striker’) or taranis (‘the thunderer’), many of these popular (and therefore presumably powerful and commanding) deities have generally been the irst victims of interpretatio: in other words, the local people referred to them by an appropriate Latin name when setting up Latin inscriptions. We are therefore predominantly dealing with ‘native’ variations of Jupiter, Juno, Mars, Mercury, Minerva, Silvanus, etc., that continued to evolve in the Principate, creating deities that were neither ‘Roman’ nor (pre-Roman) ‘Celtic’, but typical for their time, their local setting, and local society. Finally, we need to consider the wider archaeological context. Are we dealing with ‘Celtic’ or ‘RomanoCeltic’ religion when we ind votive inscriptions in a Gallo-Roman or Britanno-Roman sanctuary with nonRoman architecture and cult practices? For example, the sanctuary in Uley (Gloucestershire) is rather ‘nonRoman’ in origin and character, despite the cult statue to Mercury, and the inds from Uley ought to feature in a corpus of the fontes of Celtic religions. The archaeological evidence reveals an enormous continuity of ritual practices from the late Iron Age down to the mid4th century AD.39 At some stage during the sanctuary’s existence, somebody King 2005, 332–4. [ 17 ] wrth gwrs ‘Iau-y-Cawr-o-Farchogwr’ (Jupitergigantenreiter neu cavalier à l’anguipède), a oedd yn enwedig o nodweddiadol yn nwyrain Gâl, a dadansoddir eu perthynas â’r crefyddau Galaidd-Rufeinig gan Florian Blanchard yn y gyfrol hon. Er eu bod weithiau’n gysylltiedig ag enwau duwiau (neu deitlau) mewn iaith Geltaidd, megis sucellos (‘yr ergydiwr da’) neu taranis (‘y taranwr’), yn gyffredinol roedd nifer o’r duwiau poblogaidd (ac o’r herwydd, pwerus a llywodraethol, fwy na thebyg) gyda’r cyntaf i ddioddef oherwydd interpretatio: mewn geiriau eraill, cyfeiriodd y bobl leol atynt wrth enw Lladin priodol wrth osod arysgrifau Lladin. Felly, rydym yn ymdrin yn bennaf ag amrywiadau ‘brodorol’ ar Iau, Iwno, Mawrth Mercher, Minerva, Silvanus, ac ati, a barhaodd i esblygu yn y Dywysogaeth, gan greu duwiau nad oedd yn ‘Rhufeinig’ nac yn ‘Rhufeinig’ na ‘Cheltaidd’ (cyn-Rufeinig), ond yn nodweddiadol o’u hamser, eu cyddestun lleol a’r gymdeithas leol. Yn olaf, mae angen i ni ystyried y cyd-destun archeolegol ehangach. Ai crefydd ‘Geltaidd’ neu ‘GeltaiddRufeinig’ sydd dan sylw pan geir arysgrifen addunedol mewn cysegr Galaidd-Rufeinig neu BrythonigRufeinig gyda phensaernïaeth anRufeinig ac arferion defodol? Er enghraifft, nid yw’r cysegr yn Uley (Swydd Caerloyw) yn ‘Rhufeinig’ iawn ei darddiad na’i nodweddion, er bod cer lun addoli i Mercher, a dylai canfyddiadau Uley fod yn rhan o gorpws fontes crefyddau Celtaidd. Mae’r dystiolaeth archeolegol yn datgelu parhad nifer helaeth o arferion defodol o ddiwedd yr Oes Haearn i [ 18 ] 40 41 42 I. c e lt ic r e l igions in th e Rom a n pe r io d irst made the association of the local deity with Mercury and somebody else, perhaps when the Romano-Celtic temples was constructed, set up a statue of Mercury. These individual acts set in motion long-term processes and we can presume that in the long term this was probably changing people’s understanding of their local god: a hybridised, local deity develops that taken up additional features, while other aspects of his cult or myth might have become forgotten, even in oral tradition. Uley, like other RomanoCeltic sanctuaries, also shows how writing could be adopted and adapted to serve an ‘autochthonous’ cult. Here, the use of curse tablets, similar to Bath, but different from many continental de ixiones (cf. Roger Tomlin’s paper), could perhaps be an expression of the superstitio of the local people that Romans remarked upon. In other sanctuaries, like Glanum/Glanon in Southern Gaul, writing had already been adopted in the late Iron Age, and one Gallo-Greek inscription mentions the god Belenos.40 But his name is no longer attested in Roman Glanum. Instead, Hercules was worshipped at the sacred spring, as a healing god:41 did he replace Belenos here, rather than the commonly assumed counterpart, Apollo (cf. paper by Ralph Häussler in this volume)?42 All RIG G-63. For Glanum, and the role of Hercules in particular, cf. Roth Conges 1997. Apollo Belenos/Belinus is attested at Bardonecchia (v. supra) and notably at Aquileia in the Roman period. Belenos has been traditionally translated as ‘brilliant’ or ‘luminous’, but X. Delamarre (2003, 71–2, s.v. belo-) has convincingly shown that this name is likely to 39 40 41 42 ganol y 4edd ganrif OC.39 Ar ryw adeg yn ystod bodolaeth y cysegr, gwnaeth rhywun y cysylltiad rhwng duw lleol a Mercher a chododd rhywun arall, efallai adeg adeiladu’r temlau Rhufeinig-Geltaidd, ger lun Mercher. Cychwynnodd y gweithredoedd hyn brosesau hirdymor a gellir rhagdybio mai’r tebygrwydd oedd bod hyn yn newid dealltwriaeth pobl o’u duw lleol: mae duw lleol ar ffurf hybrid yn datblygu nodweddion ychwanegol, wrth i agweddau eraill ar ei addoliad neu ei fyth gael eu hangho io, o bosibl, hyd yn oed mewn traddodiad llafar. Mae Uley, yn yr un modd ag addoliad Rhufeinig-Geltaidd arall, yn dangos sut y gellid mabwysiadu ysgrifennu a’i addasu at ddiben addoliad ‘brodorol’. Yma, gallai defnydd llechi melltithio, tebyg i Gaerfaddon, ond yn wahanol i lawer o de ixiones (cf. papur Roger Tomlin), fod yn arwydd o superstitio y bobl leol y gwnaeth y Rhufeiniaid sylw arno. Mewn cysegrau eraill, megis Glanum/Glanon yn ne Gâl, mabwysiadwyd ysgrifennu eisoes tua diwedd yr Oes Haearn, a chyfeiria un arysgrifen Galaidd-Roegaidd at y duw Belenos.40 Ond bellach nid oes sôn amdano yn Glanum Rhufeinig. Yn hytrach, addolwyd Ercwlff yn y ffynnon gysegredig honno, yn dduw iacháu: 41 a ddisodlodd Belenos yma, yn hytrach na’r duw cyfatebol a ragdybir yn gyffredinol, sef Apolon (cf. papur gan Ralph Häussler yn y gyfrol hon)?42 King 2005, 332–4. RIG G-63. Ar gyfer Glanum, a rôl Ercwlff yn enwedig, cf. Roth Conges 1997. Mae tystiolaeth i Apolon Belenos/Belinus yn Bardonecchia (v. supra) ac yn enwedig yn Acwileia yn y cyfnod Rhufeinig. Yn draddodiadol, cy ieithwyd Belenos yn ‘ddisglair’ neu Hae ussle r & King this shows the all-pervading problem of translatability between different cultures and religions. We therefore need to ask whether we are really having thousands of ‘Celtic’ gods or whether people just employed different names and labels for the same deity or divine concept, emphasising particular aspects of a local divine embodiment that was important to them. It is feasible that the same goddess was called ‘Mother (Earth)’ (mater, terra mater), ‘The Purveyer’ (rosmerta), ‘The Queen’ (regina, perhaps also Juno Regina), ‘The Well-leading (Goddess)’ (sulevia), ‘The Very Powerful (goddess)’ (belisama), or ‘of the sacred grove’ (nemetona), while other people preferred to make a comparison with Graeco-Roman deities in their Latin dedications, like Minerva, Juno Regina, Maia, Letô, and Fortuna, and perhaps also Bona Dea or Magna Mater. We ind similar phenomena in many cultures and religions; also, god’s ‘real’ name might have become a taboo name, prompting the creation of a range of alternative designations to address a deity. Others’ name of choice may relate to the locality, for example by using a particular topographical epithet, such as perhaps Sulevia Edennica Minerva, ‘the well-leading Minerva of Eyssène’.43 And sometimes we seem 43 mean ‘Maître de la Puissance’. Though Hercules and Silvanus/Sucellos clearly dominate Glanum’s epigraphic record, there is some rather uncertain evidence for Apollo: CIL XII 99 and IGF 51. From Collias (Gard), discovered at the L’Ermitage-de-Laval Chapel: CIL XII 2974 (add. 43 p. 832) = ILGN 398 = ILS 4662 = CAG 30/2, no 085, 8*; Delamarre 2003. [ 19 ] Y cyfan y mae hyn yn ei ddangos yw problem hollbresennol trosi rhwng gwahanol ddiwylliannau a chrefyddau. Felly, mae angen i ni ofyn a yw’n wir bod gennym iloedd o dduwiau ‘Celtaidd’ neu a oedd pobl yn defnyddio gwahanol enwau a labeli ar gyfer yr un duw neu gysyniad dwyfol, gan bwysleisio agweddau penodol ar ymgorfforiad dwyfol lleol a oedd yn bwysig iddynt. Mae’n bosibl y cyfeirid at yr un dduwies o dan yr enw ‘Mam(-Ddaear)’ (mater, terra mater), ‘Y Darparwr’ (rosmerta), ‘Y Frenhines’ (regina, efallai hefyd Juno Regina), ‘Y (dduwies) arweiniol’ (sulevia), ‘Y (dduwies) dra phwerus’ (belisama), neu ‘o’r gelli sanctaidd’ (nemetona), wrth i eraill ddewis gwneud cymhariaeth â duwiau GroegaiddRufeinig yn eu cysegriadau Lladin, megis Minerva, Juno Regina, Maia, Letô, a Fortuna, ac efallai hefyd Bona Dea neu Magna Mater. Cawn ni ffenomena tebyg mwn nifer o ddiwylliannau a chrefyddau; hefyd, gallai enw ‘go iawn’ fod yn enw tabŵ, gan ysgogi creu ystod o ddynodiadau amgen wrth gyfarch duw. Gallai’r enw a ddewisa eraill fod yn gysylltiedig â’r ardal, er enghraifft trwy ddefnyddio teitl topograffaidd penodol, megis, o bosibl, Sulevia Edennica Minerva, ‘Minerva ffynnonarweiniol Eyssène’.43 Ac weithiau, ‘oleuol’, ond mae X. Delamarre (2003, 71–2, s.v. belo-) wedi dangos yn argyhoeddiadol ei bod hi’n debyg mai ystyr yr enw hwn yw ‘Maître de la Puissance’. Er ei bod yn debyg mai Ercwlff a Silvanus/Sucellos sy’n cael y prif sylw yng nghofnod epigraf ig Glanum, mae peth tystiolaeth ansicr ar gyfer Apolon: CIL XII 99 ac IGF 51. O Collias (Gard), a ddarganfuwyd yng Nghapel L’Ermitage-de-Laval: CIL XII 2974 (add. t. 832) = ILGN 398 = ILS 4662 = CAG 30/2, no 085, 8*; [ 20 ] 44 45 I. c e lt ic r e l igions in th e Rom a n pe r io d to be dealing with a personi ication, either of a particular function (similar to the Roman Fortuna, Abundantia, Ops, Pax, Salus, etc.), or of a town, river, or mountain. We need to be aware that opposing interpretations are frequently possible, like a deonomastic and a functional one: for example, the goddess Segomanna is traditionally associated with the river Seynes, but she could also have been a supraregional, powerful goddess: ‘Greatby-her-Victories’.44 Certain male names, too, like ‘the Divine’ (divanno), ‘the All-mighty’ (lanovalus),45 ‘the Master of Force’ (belenos), or the ‘Powerful’ (vernostonos), seem rather interchangeable and non-speci ic, while ‘surnames’ like ‘the thunderer’ (taranis) or ’the good striker’ (sucellos) are already less interchangeable as they relate to a deity’s particular function or myth. But none of them is necessarily the ‘real’ name of a god. And then there are of course the eight Celtic epithets (e.g., Andicrosos, Ausecos, Medugenos) that served to convey the meaning of a god called Faunus from the 4th-century hoard discovered near the then abandoned Iron Age sanctuary of Thetford; in her paper, Daphne Nash-Briggs also argues for ‘speculative etymologizing, verbal and visual riddles, and cryptic allusions’ in the religious sphere, i.e. an educated elite who was capable to make up new theonyms and epithets. E.g., ILGN 393; ILS 9311; CAG 30/3, no 319, 1*, 44 p. 686; CAG-30/2, p. 353; for the etymology of Segomanna, cf. Delamarre 2003, 268, s.v. sego‘victoire, force’: ‘Grand-par-ses-Victoires’. Delamarre 2003, 196, 305 for Lanovalus as 45 ‘Tout-Puissant’ or ‘Plein Prince’. ceir yr argraff mai’r hyn sydd gennym yw personoli, boed yn swyddogaeth benodol (debyg i Fortuna, Abundantia, Ops, Pax, Salus Rhufain ac ati), neu’n dref, yn afon, neu’n fynydd. Mae angen i ni fod yn ymwybodol bod dehongliadau cyferbyniol yn aml yn bosibl, megis un deonomastig ac un swyddogaethol: er enghraifft, cysylltir y dduwies Segomana yn aml ag afon Seynes, ond gallai hefyd fod yn dduwies bwerus, uwch-ranba rthol: ‘Mawr-trwy-eiChampau’.44 Ymddengys fod rhai enwau, dwyfol, hefyd, megis ‘y Dwyfol’ (divanno), ‘yr Holl-alluog’ (lanovalus),45 ‘Meistr Grym’ (belenos), neu’r ‘Pwerus’ (vernostonos), braidd yn gyfnewidiol ac amhenodol, a ‘chyfenwau’ megis ‘y taranwr’ (taranis) neu’r ’ergydiwr da’ (sucellos) eisoes yn llai cyfnewidiol am eu bod yn ymwneud â swyddogaeth neu fyth penodol y duw. Ond nid enw ‘go iawn’ rhyw dduw yw’r un o’r rhain o reidrwydd. Ac wedyn, wrth gwrs, yr wyth teitl Celtaidd (e.e. Andicrosos, Ausecos, Medugenos) a oedd yn fodd i gy leu ystyr o’r enw Faunus o gelc o’r 4edd ganrif a ddarganfuwyd ger cysegr o’r Oes Haearn yn Thetford a oedd wedi ei adael erbyn hynny; yn ei phapur, dadleua Daphne Nash-Briggs hefyd am ‘etymoleiddio dyfaliadol, posau llafar a gweledol, a chyfeiriadau cryptig’ yn y maes crefyddol, e.e. elit addysgedig a oedd yn gallu bathu enwau duwiau a theitlau newydd. Delamarre 2003. E.e., ILGN 393; ILS 9311; CAG 30/3, no 319, 1*, t. 686; CAG-30/2, t. 353; am darddiad Segomanna, cf. Delamarre 2003, 268, s.v. sego‘victoire, force’: ‘Grand-par-ses-Victoires’. Delamarre 2003, 196, 305 ar gyfer Lanovalus yn ‘Tout-Puissant’ neu ‘Plein Prince’. Hae ussle r & King [ 21 ] Local, personal, global Lleol, personol, byd-eang Signi icant for the advancements since 1998 have been three crucial aspects of the F.E.R.C.AN. Project: its diachronic perspective, the pan-European viewpoint, and the multidisciplinary approach, involving inter alia archaeologists, historians, philologists, and linguists from across Europe. Among others, this has exposed both striking parallels and profound divergences in our evidence across the Celtic world, both in the Iron Age and during the subsequent transformation of cults in the Roman period. This recognition already prompted a minor change of the project’s title ten years ago from ‘Celtic’ religion to religions since we are certainly dealing, despite all super icial resemblances, with rather localised religious understandings and cult practices, and it seems hardly feasible that they all derive from a ‘common proto-Celtic religion’ from which all Iron Age and Roman cults derive; many of the parallels between Celtic regions do not necessarily attest a common ‘Celtic ancestry’, as we can also ind parallels with other Indo-European and non-Indo-European cultures. Despite certain analogies, diversity seems to dominate, suggesting that we are dealing with many localised phenomena. Consequently, we need to ask whether there ever there any pan-Celtic deities or pan-Celtic myths and believes, or whether they are all a reconstruction based on feeble evidence (see for example the paper by John Koch and Fernando Fernández Rhan arwyddocaol o’r datblygiadau er 1998 fu tair agwedd hollbwysig ar Brosiect F.E.R.C.AN.: ei bersbectif diachronig, y sa bwynt panEwropeaidd, a’r dull amlddisgyblaethol, sy’n cynnwys inter alia archeolegwyr, haneswyr, ieithegwyr ac ieithyddion o bob cwr o Ewrop. Ymhlith eraill, mae hyn wedi datgelu cyfochredd trawiadol a gwahaniaethau mawr o ran ein tystiolaeth ar draws y byd Celtaidd, yn yr Oes Haearn ac yn ystod gweddnewidiad dilynol y defodau yn ystod y cyfnod Rhufeinig. Arweiniodd y gydnabyddiaeth hon eisoes at fân newid yn nheitl y prosiect ddeng mlynedd yn ôl o’r grefydd ‘Geltaidd’ i grefyddau gan ei bod yn sicr ein bod yn ymwneud, er gwaethaf pob tebygrwydd arwynebol, â dealltwriaeth grefyddol leol ac arferion addoli, a go brin fod pob un yn deillio o ‘grefydd gyffredin broto-Geltaidd’ sydd wrth wraidd holl ddefodau’r Oes Haearn a’r Cyfnod Rhufeinig: nid yw nifer o’r adleisiau rhwng rhanbarthau Celtaidd o reidrwydd yn brawf o ‘linach Geltaidd’ gyffredin, mae adleisiau i’w cael ymhlith diwylliannau IndoEwropeaidd ac an-Indo-Ewropeaidd eraill. Er gwaethaf rhai cymariaethau, ymddengys fod amrywiaeth wedi mynd â hi, gan awgrymu ein bod ni’n ymwneud â llawer o ffenomenâu. O ganlyniad, mae angen i ni holi a fu unrhyw dduwiau pan-Geltaidd neu fythau a chredoau pan-Geltaidd, neu a yw pob yn ymgais i ail-greu ar sail tystiolaeth wan (gweler, er enghraifft, y papur gan J. Koch ac F. Fernández am astudiaeth i’r duw pan-Geltaidd Lugus [ 22 ] 46 47 48 I. c e lt ic r e l igions in th e Rom a n pe r io d for a study of the elusive pan-Celtic god Lugus who gave rise to so many toponyms, notably Lugdunum/Lyon).46 This leads to the theme of this volume: Celtic religions: local, personal, global. On the one hand, we can see certain features that seem to reappear across the Keltiké, such as theonyms like Epona, Esus, Rosmerta, and Toutatis, though we should not ignore subtle (and sometimes less subtle) differences in time and space.47 So-called ‘Celtic’ religions, like contemporary Greek and Roman cults, always seem to have been tied to the locality and integrated into the local landscape: it is the religion of the local ethnos (‘polis’ or ‘tribe’), and many cults and myths were connected to local springs, rivers, lakes, and mountains, often connecting and adapting mythical accounts to particular locations (cf. Marjeta Šašel Kos’ paper on river deities). This helps to explain the myriad of local deities: how many hundreds of unique Celtic theonyms/epithets are known, and how many orthographic variations?48 For Lugus in general and for the various towns 46 called Lugdunum, cf. the study by Hily 2007. And we should not ignore the possibility that some ‘theonyms’ are merely surnames, epithets, epicleses, and not the actual name of a 47 god – v. infra. The best overview is still Jufer & Luginbühl 2001. For orthographic variations, cf. for example divine names like Abianus – Abianius – Avi- 48 anus, Accorus – Adcorus, Esus – Aesos, Belenos – Belinus; Belatucadros – Baliticaurus – Belautairus, etc., and there are of course the multiple variations for the enigmatic ‘Veteres’ on Hadrian’s Wall, attested both as singular and plural (e.g., Vitiribus, Hvitiribus, Vetiri, Vetri, Vitire, Votrim, Vheteri, etc.), perhaps re lecting different attempts by individuals to render the pronunciation of a name as they perceived it sy’n llawn dirgelwch a oedd wrth wraidd cynifer o enwau lleoedd, yn enwedig Lugdunum/Lyon).46 Arweinia hyn at thema’r gyfrol hon: Crefyddau Celtaidd: lleol, personol, byd-eang. Ar y naill law, gallwn weld rhai nodweddion sy’n ymddangos eu bod yn ailymddangos ar draws y Keltiké, er enghraifft, enwau duwiau megis Epona, Esus, Rosmerta, a Touatis, er na ddylem anwybyddu mân wahaniaethau (ac weithiau rhai llai mân) o ran amser a lle.47 Ymddengys fod crefyddau ‘Celtaidd’, yn yr un modd â defodau Groegaidd a Rhufeinig, bob amser yn gysylltiedig â’r ardal leol ac yn rhan annatod o’r dirwedd leol: crefydd yr ethnos (‘polis’ neu ‘lwyth’) lleol ydyw, ac roedd nifer o ddefodau a mythau’n gysylltiedig â ffynhonnau, afonydd, llynnoedd a mynyddoedd lleol, sydd yn aml yn cysylltu ac yn addasu adroddiadau chwedlonol i leoliadau penodol (cf. Marjeta Šašel Kos’ papur ar dduwiau afon). Helpa hyn i esbonio’r myrdd o dduwiau lleol: faint o gannoedd o enwau duwiau/ teitlau Celtaidd unigryw sy’n hysbys, a faint o amrywiadau orthograf ig?48 Yn Ar gyfer Lugus yn gyffredinol a’r gwahanol drei o’r enw Lugdunum, cf. yr astudiaeth gan Hily 2007. Ac ni ddylem anwybyddu’r posibilrwydd mai llysenwau, teitlau, arddeisy iadau yn unig yw rhai ‘enwau duwiau’ yn hytrach nag enw go iawn y duw – v. infra. Y trosolwg gorau o hyd yw Jufer a Luginbühl 2001. Ar gyfer amrywiadau orthograf ig, cf. enwau dwyfol megis Abianus – Abianius – Avianus, Accorus – Adcorus, Esus – Aesos, Belenos – Belinus; Belatucadros – Baliticaurus – Belautairus, ac ati, ac mae, wrth gwrs, llu o amrywiadau ar gyfer y term enigmatig ‘Veteres’ ar Fur Hadrian, a geir ar ffurf unigol a lluosog (e.e., Vitiribus, Hvitiribus, Vetiri, Vetri, Vitire, Votrim, Vheteri, ac ati), gan adlewyrchu, o bosibl, ym- Hae ussle r & King In addition, in Roman times, we need to take into account how societal and cultural developments across the empire must have in luenced local religious understanding.49 This led to local people rede ining their cults and cult practices, not just vis-à-vis GraecoRoman cults, but also in response to Graeco-Oriental ‘mystery’ cults that spread throughout the Keltiké, challenging people’s traditional religious understandings. This also leads us to the theme of personal religion: certainly in the Roman empire, the individual social agent could make personal choices in the sphere of religion, more than ever before (accelerated by staggering migration, societal complexity, and increased cultural interaction). It is also these individuals that are responsible for the bulk of our sources: the countless votive inscriptions. Addressing a deity by a Celtic, Roman, or Greek name, choosing a particular form of ex-voto, or anthropomorphic representation seems to have increasingly become a personal choice that ran parallel to and complemented the civic cults of the local community: how many cult places can be found in suburban and rural locations, frequently of minuscule size, far away from the sway of the local ordo?50 49 50 into Latin characters; cf. Birley 1980, 107–8; Jufer & Luginbühl 2001, 71–3. Cf. papers on continuity and innovation between Iron Age and Roman period in Haeussler and King (edd.) 2007–2008. Cf. Haeussler 2008a; 2014b for examples of numerous small-scale cult places in Southern Gaul, like Lioux, Les Milles (Aix-en-Provence), 49 and many more; for ‘peri-urban’ cult places of varying size and monumentality, cf. the study [ 23 ] ogystal, adeg y Rhufeiniaid, mae angen i ni gymryd i ystyriaeth y modd y mae’n rhaid bod datblygiadau cymdeithasol a diwylliannol ar draws yr ymerodraeth wedi dylanwadu ar ddealltwriaeth grefyddol leol.49 Arweiniodd hyn at bobl leol yn ailddif inio eu defodau a’u dulliau addoli, nid yn unig ynghylch defodau Groegaidd-Rufeinig, ond hefyd mewn defodau ‘dirgelwch’ GroegaiddDdwyreiniol a ledodd ar draws y Keltiké, gan herio dehongliadau crefyddol traddodiadol pobl. Mae hyn hefyd yn ein harwain ni at thema crefydd bersonol: yn sicr, yn yr ymerodraeth Rufeinig, gallai’r asiant cymdeithasol unigol wneud dewisiadau personol yng nghylch crefydd, yn fwy nag erioed o’r blaen (rhywbeth a gy lymwyd trwy allfudo graddol, cymhlethdod cymdeithasol, a chynnydd o ran rhyngweithio diwylliannol). Yr unigolion hyn hefyd sy’n gyfrifol am y rhan fwyaf o’n ffynonellau: yr arysgrifau addunedol diddiwedd. Ymddengys fod cyfarch duw wrth enw Celtaidd, Rhufeinig neu Roegaidd, dewis ffurf arbennig ar offrymau llw, neu gynrychiolaeth anthropomorf ig wedi mynd yn fwy ac yn fwy yn ddewis personol a oedd yn cydredeg â defodau dinesig y gymuned leol ac yn ategu’r defodau hynny: faint o leoedd addoli y gellir dod o hyd iddynt mewn maestre i a lleoliadau gwledig, yn aml o faint bach, bach ymhell i drechion gan unigolion i gy leu ynganiad enw fel y’i clywsant i gymeriadau Lladin; cf. Birley 1980, 107–8; Jufer a Luginbühl 2001, 71–3. Cf. papurau ar barhad ac arloesi rhwng cyfnod yr Oes Haearn a’r cyfnod Rhufeinig yn Haeussler a King (gol.) 2007–2008. [ 24 ] 51 52 I. c e lt ic r e l igions in th e Rom a n pe r io d This personal religion is nowhere more obvious, and more enigmatic, than in the case of Hadrian’s Wall: while the army calendar prescribed certain sacri ices to emperors and festivals for Roman state gods,51 the individual soldier must have been free to worship whatever deity he wished, resulting in countless altars to deities bearing Celtic names, like Belatucadros, Cocidios, Veteres, and many more (but of course also deities like Mithras). But are we dealing with autochthonous deities or imported deities from the Continent? Imported deities are more obvious in the case of ‘Germanic’ theonyms, like Thincsus, worshipped by people who frequently identi ied themselves as coming from Germania Inferior.52 But who, for example, was by Pechoux 2010. 50 As in the case of the Feriale Duranum: P.Dura 54; cf. Herz 1998. A very interesting example for this Germanic diaspora identity comes from Housesteads on Hadrian’s Wall; set up by the Germani from Twent, they not only worship Mars Thinc- 51 sus, but also the ‘two Alaisiagae’: Deo | Marti | Thincso | et duabus | Alaisiagis | Bed(a)e 52 et Fi|mmilen(a)e | et n(umini) Aug(usti) Ger|m(ani) cives Tu|ihanti | v(otum) s(olverunt) l(ibentes) m(erito) (RIB 1593 = CSIR-GB I.6, 159 = ILS 4760). In a similar dedication from Housesteads, the same Germani cives Tuihanti (this time additionally identi ied as cunei Frisiorum Ver(covicianoum)) only made a dedication to Mars and the two Alaisiagae (RIB 1594 = CSIR-GB I.6, 160 = ILS 4761): surely, they must have just left out the epithet and the same Mars Thincsus was implied; the nature of the two Alaisiagae is still problematic: probably ‘Germanic’? The situation is complex due to these Alaisiage being called Beda and Fimmilena on RIB 1593 and Baudihillia et Friagabis on RIB 1576. But what about the dedication to deus Mars by Calve() Ger(manus) at Housesteads: did he worship Mars Thincsus, Mars Cocidius, or perhaps the Roman Mars? At Ebchester/Vindomora, a certain Virilis Ger(manus) made a dedication deo Vernostono ffwrdd o batrwm yr ordo lleol?50 Nid yw’r grefydd bersonol hon yn fwy amlwg yn unman, nac yn fwy enigmatig, nag yn achos Mur Hadrian: er mai calendr y fyddin oedd yn pennu rhai aberthau i ymerawdwyr a gwyliau duwiau’r wladwriaeth Rufeinig,51 rhaid bod y milwr unigol yn rhydd i addoli pa dduw bynnag a fynnai, gan arwain at allorau ar gyfer duwiau sydd ag enwau Celtaidd, megis Belatucadros, Cocidios, Veteres, a llawer rhagor (ond hefyd duwiau megis Mithras). Ond ai duwiau brodorol sydd dan sylw neu rai a fewnforiwyd o’r Cyfandir? Mae duwiau a fewnforiwyd yn amlycach yn achos enwau duwiau ‘Germanig’, megis Thincsus, a addolwyd gan bobl a nodai yn aml eu bod yn hanu o Germania Leiaf.52 Ond pwy, er enghraifft, oedd Cf. Haeussler 2008a; 2014b am enghreifftiau o nifer o leoedd addoli ar raddfa fach yn ne Gâl, megis Lioux, Les Milles (Aix-en-Provence), a llawer rhagor; ar gyfer lleoedd addoli ‘peritrefol’ o amrywiol faint a choffadwyedd, cf. yr astudiaeth gan Pechoux 2010. Megis yn achos y Feriale Duranum: P.Dura 54; cf. Herz 1998. Daw enghraifft ddiddorol iawn o hunaniaeth y diaspora Germanig hwn o Housesteads ar Fur Hadrian; sefydlwyd y rhain gan y Germani o Twent, maent yn addoli nid yn unig Mars Thincsus, ond hefyd y ‘ddwy Alaisiaga’: Deo | Marti | Thincso | et duabus | Alaisiagis | Bed(a)e et Fi|mmilen(a)e | et n(umini) Aug(usti) Ger|m(ani) cives Tu|ihanti | v(otum) s(olverunt) l(ibentes) m(erito) (RIB 1593 = CSIR-GB I.6, 159 = ILS 4760). Mewn cysegriad tebyg o Housesteads, gwnaeth yr un Germani cives Tuihanti (y tro hwn, nodir yn ychwanegol eu bod yn cunei Frisiorum Ver(covicianoum)) gysegriad yn unig i Fawrth a’r ddwy Alaisiaga (RIB 1594 = CSIR-GB I.6, 160 = ILS 4761): rhaid eu bod wedi hepgor y teitl a’r un Mars Thincsus a olygid; erys natur y ddwy Alaisiaga yn broblem o hyd: ‘Almaenig’ fwy na thebyg? Mae’r sefyllfa yn un gymhleth am mai’r enw a roddir ar yr Alaisiagae yw Beda a Fimmilena a RIB 1593 a Baudihillia a Friagabis ar RIB 1576. Beth am y cysegriad i deus Mars gan Calve() Ger(manus) Hae ussle r & King Coventina at Carrawburgh (Brocolita, an auxiliary fortress): was she really an autochthonous ‘Romano-Celtic goddess of water and springs’, and do these inscriptions reveal a preexisting sacred site at Carrawburgh, prior to the Claudian invasion (cf. paper by Fernando Fernández)? The enigma becomes even larger when we look at Britain as a whole: whilst most ‘Celtic theonyms’ are attested on inscriptions from the North, in the south we ind a large number of sanctuaries that were in use from the late Iron Age into the Principate: cult practices there often only evolved gradually, suggesting an enormous degree of religious ‘persistence’,53 but interestingly most of these sanctuaries, where cult practices hardly changed over many generations, usually yield no inscription at all (see for example 53 Cocidio (RIB 1102), using a ‘Celtic’ formula; and at Brampton, the Germani Duio, Ramio, Trupo and Lurio ful illed their vow to a god with a Celtic name, deus Maponus (RIB 2063 – CSIR GB I.6, 158 = ILS 4640); Labareus Ge(rmanus) worshipped dea Setlocenia at Maryport/Alauna (RIB 841), and Aurelius Crotus German(us) and Maduhus Germ(anus) worshipped Coventina: die(!) Coventine(!) and dea nimfa(!) coventine(!) respectively, at Carrawburgh/Brocolita (RIB 1525–6 = CSIR GB I.6, 144–5) (for Coventina, see paper by Fernando Fernández in this volume). Those ‘Germans’ equally worshipped established cults in Roman Britain and their own, imported deities, like Thincus and the matres Germaniae (worshipped by M(arcus) Senec[ia] nius V[---] at Housesteads, RIB 652; the vex(illatio) Germa[no]r(um) V[o]r[e]d(ensium) worshipped the deae matrae transmarinis at Old Penrith/Voreda: RIB 920). For example, bone evidence shows an enor- 53 mous degree of continuity over generations, even centuries, from the late Iron Age (1st century BC) well into the post-invasion period, sometimes up to the 2nd–3rd century AD; cf. King 2005. [ 25 ] Coventina yn Carrawburgh (Brocolita, caer gynorthwyol): ai duwies frodorol Rufeinig-Geltaidd dŵr a ffynhonnau oedd hi, ac a yw’r arysgrifau hyn yn datgelu sa le cysegredig blaenorol yn Carrawburgh cyn yr ymosodiad o dan Claudius (cf. papur gan F. Fernández)? Cynyddu a wna’r dirgelwch pan edrychwn ni ar Brydain yn ei chyfanrwydd: er bod tystiolaeth i’r rhan fwyaf o ‘enwau duwiau Celtaidd’ ar arysgrifau o’r Gogledd, yn y de gwelir bod nifer fawr o gysegrau yn cael eu defnyddio o ddiwedd yr Oes Haearn hyd y Dywysogaeth: yn aml esblygu yn raddol yn unig a wna’r arferion addoli, gan awgrymu gradd fawr o ‘barhad’ crefyddol,53 ond yn ddiddorol, yn achos y rhan fwyaf o’r cysegrau hyn, fel arfer yn Housesteads: a addolai Mars Thincsus, Mars Cocidius, neu o bosibl Mawrth Rhufain? Yn Ebchester/Vindomora, gwnaeth rhyw Virilis Ger(manus) gysegriad i deo Vernostono Cocidio (RIB 1102), gan ddefnyddio fformiwla ‘Celtaidd’; ac yn Brampton, Cy lawnodd y Germani Duio, Ramio, Trupo a Lurio eu hadduned i dduw oedd ag enw Celtaidd, deus Maponus (RIB 2063 – CSIR GB I.6, 158 = ILS 4640); addolodd Labareus Ge(rmanus) dea Setlocenia ym Maryport/Alauna (RIB 841), ac addolodd Aurelius Crotus German(us) a Maduhus Germ(anus) Coventina: die(!) Coventine(!) and dea nimfa(!) coventine(!) yn eu tro, yn Carrawburgh/Brocolita (RIB 1525–6 = CSIR GB I.6, 144–5) (ar gyfer Coventina, gweler y papur gan Fernando Fernández yn y gyfrol hon). Roedd y ‘Germaniaid’ hynny yn addoli i’r un graddau defodau a oedd wedi ennill eu plwyf ym Mhrydain y Rhufeiniaid a’u duwiau eu hunain a fewnforiwyd, megis Thincus a’r matres Germaniae (a addolid gan M(arcus) Senec[ia]nius V[---] yn Housesteads, RIB 652; addolai’r vex(illatio) Germa[no] r(um) V[o]r[e]d(ensium) y deae matrae transmarinis yn Old Penrith/Voreda: RIB 920). Er enghraifft, dengys tystiolaeth yr esgyrn fod gradd fawr o barhad dros genedlaethau, hyd yn oed canrifoedd, o ddiwedd yr Oes Haearn (y ganrif gyntaf C.C.) ymhell i mewn i’r cyfnod ar ôl yr ymosodiad, weithiau hyd at y 2il–3edd ganrif O.C.; cf. King 2005. [ 26 ] I. c e lt ic r e l igions in th e Rom a n pe r io d Steven Yeates’ paper for the newly discovered cult place at Abingdon); this is not limited to rural cult places, but includes civic cults: for instance, Verulamium’s important sanctuary of Folly Lane equally revealed not a single inscription during its 200 years existence. This contrasts sharply, of course, with sanctuaries like Uley and Bath with a myriad of inscribed de ixiones (see paper by Roger Tomlin). In order to understand the epigraphic evidence, a context ual analysis is absolutely essential. How signi icant is the text of an inscription on its own? Case studies, like Glanum and Châteauneuf, serve to illustrate how crucial it is to locate inscriptions as precisely as possible within a sanctuary or town (cf. paper by Ralph Haeussler). It warns us not to associate just any ex-voto or dedication from the same locality to a particular cult, but to take more care about the separate assemblages (‘Vergesellschaftung’) that might have belonged to distinct cult places or ritual activity zones; since most inscriptions come from archaeological excavations, Glanum also allows us to explore the positioning of religious inscriptions and how people experienced them: what was visible to the general public, and what was restricted to initiates? We also need to take care of ex-votos being moved; small votive offerings, like bronze igurines (with or without inscriptions), might have been moved over long distances across the empire before being deposited as ex-voto in a local sanctuary; even small stone altars might have been moved from a ni cheir unrhyw arysgrifen o gwbl (cf. er enghraifft papur Steven Yeates ar gyfer y lle addoli sydd newydd ei ddarganfod yn Abingdon); nid yw hyn yn gyfyngedig i leoedd addoli gwledig, ond mae’n cynnwys defodau dinesig: er enghraifft, yng nghysegr pwysig Folly Lane yn Veralamium ni welwyd yr un arysgrif yn ystod 200 mlynedd ei fodolaeth. Mae hyn yn wahanol iawn, wrth gwrs, i gysegrau megis Uley a Chaerfaddon a’u myrdd o de ixiones anysgrifenedig (gweler y papur gan Roger Tomlin). Er mwyn deall y dystiolaeth epigraf ig, mae dadansoddiad cyddestunol yn gwbl anhepgor. Pa mor arwyddocaol yw testun arysgrifen ar ei ben ei hun? Mae astudiaethau achos megis Glanum a Châteauneuf, yn helpu i ddangos pa mor bwysig yw lleoli arysgrifau mor fanwl ag y bo modd mewn cysegr neu dref (cf. y papur gan Ralph Haeussler). Mae’n ein rhybuddio ni rhag cysylltu unrhyw offrwm neu gy lwyniad o’r un ardal â defod arbennig, ond i fod yn fwy gofalus ynghylch y cynulliadau ar wahân (‘Vergesellschaftung’) a oedd, o bosibl, yn perthyn i leoedd addoli neu barthau gweithgarwch defodol penodol; gan fod y rhan fwyaf o arysgrifau’n deillio o gloddiadau archeolegol. Mae Glanum hefyd yn caniatáu i ni archwilio lleoliad arysgrifen grefyddol a phro iad pobl o’r rhain: beth oedd yn weladwy i’r cyhoedd, a beth a oedd yn gyfyngedig i rai a oedd wedi eu cy lwyno i’r ddefod? Pwyll piau hi hefyd wrth ystyried offrymau llw sydd wedi eu symud; gellid bod wedi symud offrymau bach, megis f igurynnau efydd (gydag arysgrifau Hae ussle r & King 54 55 building to a seasonal makeshift cult place in the course of a year, as we can see in Anthony King’s paper. We also should not disregard standardized production, for example, of the Minerva votive leafs from the sanctuary at Ashwell which were used as ex-votos for the goddess Senuna,54 which raise so many questions: for example, why did the local people choose the image of Minerva to represent their goddess? Which aspect of Minerva did they relate to? Did they have a more intrinsic knowledge of Minerva’s myth or did her image of a powerful, seemingly bellicose goddess in armour simply appeal to them? Alternatively, we might want to see a pattern here considering that, also in Britain, in Bath/Aquae Sulis, we ind Minerva again: Sulis Minerva. While writing became an intrinsic part of certain ‘native’ sanctuaries in the Roman period, we should not forget that across the Roman West, we ind a vast number of sanctuaries of Iron Age origin where inscriptions were generally not used. And if we ind one or two inscriptions, we should question their meaning. At the (otherwise ‘anepigraphic’) sanctuary of Hayling Island, for example, the only inscription was set up by a legionary.55 Is this merely the soldier’s personal interpretatio of a local cult that had started a century prior to the Claudian invasion and only changed gradually? To what extent can such sporadic inscriptions, notably set up by Cf. Jackson and Burleigh 2007. [Na]evian[us(?)] from the legio VIII[I] – RIB III 54 55 3042. [ 27 ] neu hebddynt), dros bellteroedd maith ar draws yr ymerodraeth cyn eu gosod yn offrwm llw mewn cysegr lleol; byddai modd hefyd symud allorau cerrig bach o adeilad i le addoli tymhorol dros dro yn ystod blwyddyn, fel y gwelir ym mhapur Anthony King. Hefyd ni ddylem ddiystyru cynhyrchiad safonol, er enghraifft, dail addunedu Minerva o’r cysegr yn Ashwell a ddefnyddid yn offrymau addunedu i’r dduwies Senuna,54 sy’n codi cynifer o gwestiynau eraill: er enghraifft, pam dewisodd y bobl leol ddelw Minerva i gynrychioli eu duwies? Â pha agwedd ar Minerva oedd y rhain yn gysylltiedig? Oedd ganddyn nhw wybodaeth fwy hanfodol am chwedl Minerva neu a oedd ei delwedd hi o dduwies rymus ryfelgar mewn arfwisg at eu dant? Fel arall, efallai yr hoffem weld patrwm yma o ystyried hefyd ym Mhrydain, yng Nghaerfaddon/Aquae Sulis, ein bod yn darganfod Minerva unwaith eto: Sulis Minerva. Er i ysgrifennu fynd yn rhan annatod o rai cysegrau ‘brodorol’ yn y cyfnod Rhufeinig, ar draws y Gorllewin Rhufeinig, ni ddylid angho io y ceir nifer fawr o gysegrau o’r Oes Haearn lle na ddefnyddid arysgrifau fel rheol. Ac os ceir un neu ddwy arysgrifen, dylid cwestiynu eu hystyr. Yn y cysegr (heb epigraffeg fel arall) ar Ynys Hayling, er enghraifft, lleng ilwr oedd yr unig un i osod arysgrifen.55 Ai interpretatio personol y milwr o ddefod leol a oedd wedi dechrau ganrif cyn ymosodiad Claudius ac na newidiodd ond yn raddol, sydd dan sylw? I ba raddau Cf. Jackson a Burleigh 2007. [Na]evian[us(?)] o’r legio VIII[I] – RIB III 3042. [ 28 ] 56 I. c e lt ic r e l igions in th e Rom a n pe r io d outsiders, re lect the religious meaning of a cult? In any case, it provides an epigraphic testimony for one of the cultores that frequented a sanctuary that goes back to a pre-Roman Iron Age cult place. Then there are also unusual cases like Woodeaton: there, we do not ind any inscriptions per se, but there is a mysterious series of bronze letters that do not make up any meaningful word: were they used for ritual activities or were they attached to a cult or votive object?56 One important question remains: why did people use Celtic theonyms and epithets in otherwise (moreor-less) perfectly written Latin inscriptions? And how does this use of Celtic names change in the course of time and in the various parts of the Roman West? For example, the Celtic theonyms attested on the Parisian pilier de nautes (‘pillar of the boatmen’) during Tiberius’ reign, like Smert[ri]os, [C]ernunnos, Tarvos Trigaranus, and Esus, re lect rather different choices and modalities compared to the 2nd– 3rd century AD when the majority of votive inscriptions were set up. The Celtic theonyms from Paris are hardly ever attested. Have they all been replaced by Latin names? But what about all these diverse Celtic names of imperial times? Since it seems unlikely that all Celtic divine names were actually names of deities, then the use of these Celtic names must have served a certain purpose. First, it is feasible that these are designations or epithets that cannot be easily translated into For Woodeaton, cf. Goodchild & Kirk 1954; RIB 56 236–239e. y gall yr arysgrifau ysbeidiol hyn, yn enwedig gan ddieithriaid, adlewyrchu ystyr grefyddol addoliad? Beth bynnag, darpara dystiolaeth epigraf ig am un o’r cultores a fynychodd gysegr sy’n mynd yn ôl i le addoli Oes Haearn cynRufeinig. Mae hefyd achosion anarferol megis Woodeaton: yno nid oes unrhyw arysgrifau fel y cyfryw, ond mae cyfres ryfedd o lythrennau efydd nad ydynt yn creu unrhyw air ystyrlon: a ddefnyddid y rhain ar gyfer gweithgareddau defodol neu oeddent yn gysylltiedig â gwrthrych defodol neu offrymol?56 Erys un cwestiwn pwysig: pam y defnyddiodd pobl enwau a theitlau Celtaidd ar gyfer duwiau mewn arysgrifau Lladin ysgrifenedig graenus (fwy neu lai) fel arall? a sut mae’r defnydd hwn ar enwau Celtaidd yn newid gydag amser ac mewn amrywiol rannau o’r Gorllewin Rhufeinig? Er enghraifft, mae’r enwau Celtaidd ar dduwiau ar pilier de nautes (‘piler y cychwyr’) Paris yn ystod teyrnasiad Tiberius, megis Smert[ri]os, [C]ernunnos, Tarvos Trigaranus ac Esus, yn adlewyrchu dewisiadau braidd yn wahanol a ffur iau o’u cymharu â’r 2il a’r 3edd ganrif OC pan osodwyd y rhan fwyaf o’r arysgrifau offrwm. Does braidd dim tystiolaeth o enwau duwiau Celtaidd o Baris. A yw’r enwau Lladin wedi disodli pob un? Ond beth am bob un o’r gwahanol enwau Celtaidd hyn o adeg yr ymerodraeth? Gan ei bod yn ymddangos yn annhebygol mai enwau duwiau oedd pob enw dwyfol Celtaidd, rhaid bod yr enwau Celtaidd wedi eu defnyddio at ryw ddiben penodol. Yn Ar gyfer Wood Eaton, cf. Goodchild a Kirk 1954; RIB 236–239e. Hae ussle r & King 57 58 Latin. Is there a good Latin translation for cocidios, ‘the blood-reddened’ god?57 It is also possible that there were linguistic misunderstandings: while local people might have referred to a god, for example, as mogons, the ‘mighty, powerful (god)’, a Latinspeaking soldier, magistrate, trader, or colonist might have mistaken this as a god’s theonym, resulting in one of the many votive inscription. But within the increasingly connected and globalizing world of the Roman empire, it will also be necessary to demarcate different ‘native’ cults and deities from each other, for example by creating stronger local identities and particularities. The F.E.R.C.AN. project’s focus on the epigraphic record also meant a continuous examination and rethinking of the etymology of the various divine names, theonyms, epithets, and epicleses in Celtic. Since 1998, many linguists have scrutinised the etymology in the context of F.E.R.C.AN. workshops, like Wolfgang Meid, Xavier Delamarre, Pierre-Yves Lambert, Patrick Sims-Williams, and Patrizia de Bernardo Stempel.58 We have seen many new exciting, and sometimes controversial, translations for theonyms, like Belenos not as healing or spring god, but as ‘Maître de la Puissance’, or Sulis Minerva as ‘One-Eyed Minerva’, Iboita as ‘drinking goddess’, and the god Lucuttectos was recently suggested to be not a Celtic god If we accept P. de Bernardo Stempel’s (2008) 57 translation for Cocidios; also for Delamarre (2003, 120), Cocidius derives from cocos, coccos, ‘écarlate, rouge’. See Table 1 for the F.E.R.C.AN. proceedings. 58 [ 29 ] gyntaf, mae’n bosibl na ellir cy ieithu’r dynodiadau neu’r teitlau hyn yn hawdd i’r Lladin. Oes cy ieithiad Lladin da ar gyfer cocidos y duw ‘gwaedgoch’?57 Gall hefyd fod camddeall ieithyddol: tra byddai pobl leol wedi cyfeirio at dduw, er enghraifft, o dan yr enw mogons, y ‘pwerus, grymus (duw)’, gallai fod milwr, ynad, masnachwr, neu wladychwr Lladin ei iaith, wedi camgymryd hwn am enw’r duw, gan arwain at un o’r arysgrifau addunedol niferus. Fodd bynnag, oddi mewn i sefyllfa gynyddol gysylltiedig a byd-eang yr ymerodraeth Rufeinig, bydd hefyd angen tynnu’r f in rhwng gwahanol ddefodau a duwiau ‘brodorol’, er enghraifft, trwy greu hunaniaethau a nodweddion lleol cryfach. O ganlyniad i bwyslais prosiect F.E.R.C.AN. ar y cofnod epigraf ig, bu hefyd arholi ac ailfeddwl parhaus ar darddiad y gwahanol enwau dwyfol, enwau duwiau, teitlau, arddeisy iadau yn y Gelteg. Er 1998, mae llawer o ieithyddion wedi craffu ar darddiadau’r geiriau yng nghyd-destun gweithdai F.E.R.C.AN., megis Wolfgang Meid, Xavier Delamarre, Pierre-Yves Lambert, Patrick Sims-Williams a Patrizia de Bernardo Stempel.58 Mae gennym lawer o gy ieithiadau cyffrous newydd, ac weithiau dadleuol, ar gyfer enwau duwiau, megis Belenos nid yn dduw iacháu na gwanwyn, ond yn ‘Maître de la Puissance’, neu Sulis Minerva yn ‘Minerva Os derbynnir cy ieithiad P. de Bernardo Stempel (2008) ar gyfer Cocidios; hefyd ar gyfer Delamarre (2003, 120), daw Cocidius o cocos, coccos, ‘écarlate, rouge’. Gweler Tabl 1 ar gyfer trafodion F.E.R.C.AN. [ 30 ] 59 60 61 62 63 I. c e lt ic r e l igions in th e Rom a n pe r io d (like ‘son/descendant of Lugus’),59 but an epithet for Apollo: a ‘mice-catching Apollo’.60 Also presumed deities, like Mogons, are now considered to be mere epithets that can be attributed to different gods. The idea of a secondary theonym is interesting: epithets were increasingly used as the proper name for a god, as in the case of (Mars) Cocidios and (Apollo) Maponos.61 And inally, there has been some thought about the origin of some of these Celtic-language theonyms and epithets.62 It is important to rethink existing paradigms, but all new interpretations equally need to be scrutinised thoroughly. All this has led to a better understanding. We are able today to recognize different categories of ‘divine names’, and thus being able to recognise more complex divine formulae. An important observation relates to multiple theonyms, like Apollo Maponus or Mars Cocidios. Rather than to interpret them as ‘con lations’ of two comparable or even identical deities, a Celtic and a Roman one, many of these combinations suggest that the Celtic word is merely a kind of epithet: in this case Apollo ‘the son of god (i.e. Zeus)’ and ‘the blood-reddened’ Mars respectively.63 Another interesting example is this dedication from Noricum: Marti | Latobio | Marmogio ILN-3, 203. De Albentiis Hienz & De Bernardo Stempel 2013. Cf. De Bernardo Stempel 2008 for the term secondary theonym. Cf. e.g. De Bernardo Stempel 2007, 2008; De Albentiis Hienz & De Bernardo Stempel 2013. Following De Bernardo Stempel 2008; Delamarre 2003. 59 60 61 62 63 Unllygeidiog’, Iboita yn ‘dduwies yfed’, ac yn ddiweddar awgrymwyd nad duw Celtaidd oedd Lucuttectos (megis ‘mab/disgynnydd Lugus’),59 ond teitl ar gyfer Apolon: ‘Apolon y daliwr llygod’.60 Hefyd bellach ystyrir bod duwiau rhagdybiedig, megis Mogons, bellach i’w hystyried yn deitlau i’w priodoli i wahanol dduwiau. Mae syniad enw duw eilaidd yn ddiddorol: yn fwy ac yn fwy, defnyddid teitlau’n enw priod ar gyfer duw, megis yn achos [Mawrth] Cocidos ac [Apolon] Maponos.61 Ac yn olaf, bu peth meddwl am darddiad yr enwau duwiau a theitlau Celteg.62 Mae’n bwysig ailfeddwl paradeimau presennol, ond mae hefyd angen craffu ar yr holl ddehongliadau newydd yn drylwyr. Mae’r cyfan wedi arwain at well dealltwriaeth. Heddiw mae modd i ni gydnabod gwahanol gategorïau ‘enwau dwyfol’, a thrwy hynny y gallu i adnabod fformiwlâu dwyfol mwy cymhleth. Mae sylw pwysig yn ymwneud â nifer o enwau duwiau, megis Apolon Maponus neu Mawrth Cocidios. Yn hytrach na dehongli’r rhain yn ‘gyfuniadau’ o ddau dduw cymharol neu hyd yn oed cyfystyr, y naill yn Geltaidd a’r llall yn Rhufeinig, awgryma nifer o’r cyfuniadau hyn mai math o deitl yn unig yw’r enw Celtaidd: yn yr achos hwn, Apolon ‘mab duw (h.y. Zeus)’ a Mawrth ‘y gwaedgoch’ yn eu tro.63 Enghraifft ddiddorol ILN-3, 203. De Albentiis Hienz a De Bernardo Stempel 2013. Cf. De Bernardo Stempel 2008 gyfer y term enw duw eilaidd (secondary theonym). Cf. e.e. De Bernardo Stempel 2007, 2008; De Albentiis Hienz a De Bernardo Stempel 2013. Yn dilyn De Bernardo Stempel 2008; Delamarre 2003. Hae ussle r & King 64 65 | Toutati | Sinati Mog|[e]tio C(aius) Val(erius) | [V]alerinus | ex voto. Gaius Valerius Valerinus did not worship a long list of six deities, but in view of what we have seen so far in this paper, there are obviously at most two gods involved: Mars and Sinatis, i.e. the very mighty (marmogios) Mars Latobios (‘of the plain’?) and the mighty (mogetios) Sinatis of the people/touta.64 The case of Apollo Maponus also leads us to another important theme: the medieval Welsh and Irish literature. After all, the Britanno-Roman Maponus seems to have become Mabon in the Welsh Mabinogi; and the name Mabon, son of Mellt (i.e. ‘Lightning’), seems appropriate to identify the son of a weather god. But we cannot just create one-to-one equivalents between ancient and medieval sources. With the organisation of the XIIIth F.E.R.C.AN. Workshop in Lampeter, the meeting was hosted for the irst time in a Celticspeaking country which provided the incentive to explore in more detail the evolution of medieval Welsh and Irish myths and to critically review their usefulness for the study of Iron Age and Romano-Celtic cults (see paper by John Koch and Fernando Fernández, and for the ‘Anti-Nativism’ debate see Jonathan Wooding’s paper).65 This is important for our methodology since there are a number of similarities between our Romano-Celtic evidence and the Welsh and Irish mythologies. What is the relationship between Lugus, the Irish Lugh, and the Welsh Llew? Are CIL III 11721 = ILS 4566; cf. De Bernardo Stem- 64 pel 2005 for etymological discussion. Also cf. Wooding 2009. 65 [ 31 ] arall yw’r cysegriad hwn o Noricum: Marti | Latobio | Marmogio | Toutati | Sinati Mog|[e]tio C(aius) Val(erius) | [V]alerinus | ex voto. Nid oedd Gaius Valerius Valerinus yn addoli rhestr hir o chwe duw, ond o ystyried yr hyn yr ydym eisoes wedi ei weld yn y papur hwn, mae’n amlwg mai dau dduw sydd o dan sylw ar y mwyaf: Mawrth a Sinatis, h.y. Tra nethol (mormogis) Fawrth Latobios (‘y gwastadedd’?) a nerthol (mogetios) Sinatis y bobl/ touta.64 Mae achos Apolon Maponus hefyd yn ein harwain ni at thema bwysig arall: llenyddiaeth Gymraeg a Gwyddeleg yr Oesoedd Canol. Wedi’r cyfan, mae lle i gredu mai Maponus y Brythoniaid Rhufeinig sydd wrth wraidd Mabon yn y Mabinogi: a’r enw Mabon, mab Mellt yn ymddangos yn briodol ar gyfer dynodi mab un o dduwiau’r tywydd. Ond ni allwn ni greu cyfatebiaethau fesul un rhwng yr hen ffynonellau a rhai’r oesoedd canol. A XIIIeg Gweithdy F.E.R.C.AN. yn Llanbedr Pont Steffan, hwn oedd y tro cyntaf i’r cyfarfod gael ei gynnal mewn gwlad lle siaredir iaith a oedd yn gymhelliad i archwilio’n fanylach esblygiad chwedlau Cymraeg a Gwyddeleg yr Oesoedd Canol ac adolygu eu defnyddioldeb ar gyfer defodau’r Oes Haearn a’r cyfnod Rhufeinig-Geltaidd (Gweler y papur gan John Koch a Fernando Fernández, ac ar gyfer y ddadl ‘Gwrth-Frodoriaeth’ gweler papur Jonathan Wooding).65 Mae hwn yn bwysig i’n methodoleg gan fod nifer o bethau cyffredin rhwng CIL III 11721 = ILS 4566; cf. De Bernardo Stempel 2005 am drafodaeth etymolegol. Hefyd cf. Wooding 2009. [ 32 ] I. c e lt ic r e l igions in th e Rom a n pe r io d there any differences between Irish Samhain and the samonios on the calendar from Coligny? How useful is the Rhiannon myth from the Mabinogi to understand (and reconstruct) the Romano-Celtic goddess Epona?66 And to what extent was Welsh mythology also the product of over 350 years of Roman ‘occupation’ in Britain? It will be important to explore further the transitional period between late Antiquity and the early Middle Ages, and how, despite Christianity, ‘Celtic’ myths and cults continued to evolve; after all, certain (Britanno-Roman) cult aspects seem to have lourished down to the 4th century in Britain, judging from the Romano-Celtic temples and sanctuaries in Caerwent, Lydney Park, Maiden Castle, Thetford, and many other sites. What was their impact on early medieval institutions, practices, and folklores? We hope that this overview provides some insight into the vast methodological problems that one encounters in the study of Celtic religions and deities. The papers in this volume each re lect a large variety of methodological approaches to the topic, re lecting different disciplines and traditions and discussing a multitude of evidence. 66 Cf. e.g. Haeussler 2008b with further biblio- 66 graphy. y dystiolaeth Rufeinig-Geltaidd a’r chwedlau Cymraeg a Gwyddeleg. Beth yw’r berthynas rhwng Lugus, Lugh y Gwyddelod, a Llew’r Cymry? A oes unrhyw wahaniaethau rhwng Samhain y Gwyddelod a samonios ar y calendr o Coligny? Pa mor ddefnyddiol yw chwedl Rhiannon y Mabinogi er mwyn deall (ac ailadeiladu) y dduwies RufeinigGeltaidd Epona?66 Ac i ba raddau roedd chwedlau Cymraeg hefyd yn gynnyrch dros 350 mlynedd o reolaeth Rufeinig ym Mhrydain? Bydd yn bwysig archwilio ymhellach y cyfnod pontio rhwng diwedd cyfnod yr Hen Fyd a dechrau’r Oesoedd Canol a sut, er gwaethaf Cristnogaeth, y parhaodd y chwedlau ‘Celtaidd’ i esblygu: wedi’r cyfan, ymddengys fod rhai agweddau defodol (Brythonig-Rufeinig) wedi ffynnu tan yn 4edd ganrif ym Mhrydain, a barnu wrth y temlau a’r cysegrau Rhufeinig-Geltaidd yng Nghaerwent, Parc Lydney, Castell Maiden, Thetford a llawer o sa leoedd eraill. Beth oedd eu heffaith neu ar sefydliadau, arferion, a chwedlau cynnar yr oesoedd canol? Gobeithio y bydd y trosolwg hwn yn rhoi rhyw gipolwg ar y problemau methodolegol enfawr sy’n dod i’n rhan ni wrth astudio’r crefyddau a’r duwiau Celtaidd. Mae pob un o bapurau’r gyfrol hon yn adlewyrchu amrywiaeth fawr o ddulliau methodolegol at y pwnc, gan adlewyrchu’r gwahanol ddisgyblaethau a thraddodiadau a thrafod llu o dystiolaeth. Cf. e.e. Haeussler 2008b gyda llyfryddiaeth bellach. Hae ussle r & King [ 33 ] BI B LI O GR A PHY — LL Y FR YD D I A E TH Abascal, J. M. 2002 ‘Téseras y monedas. Iconogra ía zoomorfa y formas jurídicas de la Celtiberia’, Palaeohispanica 2, 9–35. Bauchhenß, G. 2008 ‘Hercules in Gallien: facts and iction’, in: Haeussler and King (eds) 2008, vol. 2, 91–102. Birley, A. R. 1980 The People of Roman Britain, London: Batsford. Buisson A. & Abry J.-H. (eds.) 1993 Les tablettes astrologiques de Grand (Vosges) et l’astrologie en Gaule romaine, Actes de la table ronde du 18 mars 1992, Université de Lyon III. Paris, De Boccard (Coll. du Centre d’études romaines et gallo-romaines, nouvelle série, 12). De Albentiis Hienz, M. & P. de Bernardo Stempel 2013 ‘Apolls Epitheta – griechisch, lateinisch, keltisch bzw. keltorömisch. Eine Typologie der Beinamen klassicher Gottheiten‘. Geistes-, sozial- und kulturwissenschaftlicher Anzeiger 148 (1-2), 7–126. De Bernardo Stempel. P. 2005 ‘Die in Noricum belegten Gottheiten und die römisch-keltische Widmung aus Schloß Seggau‘,Keltischer Götter im Römischen Reich, ed. W. Spickermann & R. Wiegels, 15–28. Möhnesee, Bibliopolis. De Bernardo Stempel, P. 2007 ‘Einheimische keltische und keltisierte Gottheiten der Narbonensis im Vergleich‘,Auf den Spuren keltischer Götterverehrung. Akten des 5. F.E.R.C.AN.-Workshop, Graz 9.–12. Oktober 2003, ed. M. Hainzmann, 67–80. Wien, Österreichische Akademie der Wissenschaften Wien (Mitteilungen der Prähistorischen Kommission, volume 64). De Bernardo Stempel, P. 2008a ‘Strato teonimici nelle provincie romane (con esempi prevalentemente aquitani)’. In D’Encarnação, J. (ed.) 2008, 145–50. De Bernardo Stempel, P. 2008b ‘Continuity, translatio and identi icatio in Romano-Celtic religion: The Case of Britain‘, in Haeussler and King (eds) 2007, vol. 2, 67–82. De Bernardo Stempel, P. 2011 ‘Il testo pregallico della stele di Vercelli’, Finem dare: il con ine, tra sacro, profano e immaginario. A margine della stele bilingue del Museo Leone di Vercelli. Convegno internazionale, 22–24 maggio 2008, ed. G. Cantino-Wataghin, A. Rosso & F. M. Gambari, 67–79. Vercelli. Delamarre, X. 2003 Dictionnaire de la langue gauloise (2nd revised edition). Paris, Éditions Errance. Delamarre, X. 2013 ‘La structuration verticale de l’espace chez les Anciens Celtes et les déesses rhénanes Matronae Andrusteihae’, Théonymie celtique, cultes, interpretatio / Keltische Theonymie, Kulte, interpretatio. A. Hofeneder & P. de Bernardo Stempel (eds), 97–9. Wien, Österreichische Akademie der Wissenschaften. D’Encarnação, J. (ed.) 2008 Divindades indígenas em análise. Divinités pré-romaines - bilan et perspectives d’une recherche. Actas do VII workshop F.E.R.C.AN., Cascais, 25-27.05.2006 Cascais. Coimbra/Porto. Duval, P.-M. 1989. ‘Teutates, Esus, Taranis’ In: Travaux sur la Gaule (1946-1986), 275–87. Rome, École Française de Rome (Publications de l’École française de Rome, 116). Fauduet, I. 2010 Les Temples de tradition celtique en Gaule romaine (2nd revised edition). Paris, Editions Errance. Goodchild, R. & J. R. Kirk 1954 ‘The Romano Celtic Temple at Woodeaton’, Oxoniensia 19, 15–37. Gorrochategui, J. 2008 ‘Hacia el corpus de divinidades indígenas de la Novempopulana’, in: D’Encarnação, J. (ed.) 2008, 272–3. Gorrochategui, J. 2013 ‘Linguistisque et peuplement en Aquitania’. L’âge du Fer en Aquitaine et sur ses marges. Mobilité des hommes, diffusion des idées, circulation des biens dans l’espace européen à l’âge du Fer. Actes du 35e Colloque international de l’AFEAF (Bordeaux, 2–5 juin 2011), ed. A. Colin & F. Verdin, 17–32. Bordeaux (Aquitania Supplément 30). Green, M. 1992 Dictionary of Celtic myth and legend. London, Thames and Hudson. Gury, F. 2012 ‘Mars et le taureau: à propos du bloc sculpté provenant du sanctuaire de Mars Mullo à Allones (Sarthes)’. Mediterraneo antico 15/1–2, 175–98. Jackson, R. & G. Burleigh 2007 ‘The Senuna treasure and shrine at Ashwell (Herts)’, in Haeussler & King (eds), vol. 1, 37–54. Haeussler, R. & A. C. King (eds) 2007–2008 Continuity and Innovation in Religion in the Roman West. 2 vols., Portsmouth, RI, Journal of Roman Archaeology (Supplementary Series 67). Haeussler, R. & A. C. King 2007 ‘Introduction: The formation of Romano-Celtic religion(s)’. In Haeussler and King (eds) 2007, vol. 1, 7–12. Haeussler, R. 2008a ‘Pouvoir et religion dans un [ 34 ] I. c e lt ic r e l igions in th e Rom a n pe r io d paysage gallo-romain: les cités d’Apt et d’Aixen-Provence’, Romanisation et épigraphie. Études interdisciplinaires sur l’acculturation et l’identité dans l’Empire romain, ed. R. Haeussler, 155–248. Montagnac, Éditions Monique Mergoil (Archéologie et Histoire Romaine, 17). Haeussler, R. 2008b. ‘How to identify Celtic religion(s) in Roman Britain and Gaul’, in: D’Encarnação (ed.) 2008, 13–63. Haeussler, R. 2008c. ‘A new sacred landscape at the fringes of the Roman Empire: the civitas Vangionum’, in: Haeussler & King (edd.) 2008, vol. 2, 185–216. Haeussler, R. 2012 ‘Interpretatio indigena. Re-inventing local cults in a global world’, Mediterraneo Antico 15 (1–2), 143–74. Haeussler, R. 2014 ‘Manipulating the past. Rethinking Graeco-Roman accounts on “Celtic” religion’. Fraude, mentiras y engaños en el mundo antiguo, ed. F. Marco Simón, F. Pina Polo & J. Remesal Rodríguez, 35–54. Barcelona, Edicions de la universitat de Barcelona. Haeussler, R. 2014b ‘Differences in the epigraphic habit in the rural landscapes of Gallia Narbonensis’, Öffentlichkeit – Monument – Text. XIV Congressus Internationalis Epigraphiae Graecae et Latinae 27–31 Augusti MMXII, ed. W. Eck, P. Funke & M. Dohnicht, 323–45. Berlin, de Gruyer. Haeussler, R. 2015 ‘A landscape of resistance? Cults and sacred landscapes in Western Cisalpine Gaul’, Trans Padum … Vsque Ad Alpes. Roma tra il Po e le Alpi: dalla romanizzazione alla romanità. Atti del convegno Venezia 13–15 maggio 2014 (Studi e ricerche sulla Gallia Cisalpine, 26), ed. G. Cresci Marrone, 261–86. Roma, Edizione Quasar. Hainzmann, M. & P. de Bernardo Stempel 2013 ‘Interpretatio Romana vel indigena im Spiegel der Götterformulare’, Théonymie celtique, cultes, interpretatio / Keltische Theonymie, Kulte, interpretatio: X. workshop F.E.R.C.AN., Paris 24.– 26.Mai 2010, ed. A. Hofeneder & P. de Bernardo Stempel, 193–220. Wien: Österreichische Akademie der Wissenschaften. Haselgrove, C. 1995. ‘Social and symbolic order in the origins and layout of Roman villas in Northern Gaul’, Integration in the early Roman West. The role of culture and ideology, ed. J. Metzler, M. Millett, N. Roymans, & J. Slofstra, 65–76. Luxembourg, Dossiers d’Archéologie du Musée d’Histoire et d’Art IV. Herz, P. 1998 ‘Feriale Duranum’, Der Neue Pauly IV, 480–1. Hily, G. 2007 Le dieu celtique Lugus (PhD thesis). Paris, Ecole pratique des hautes etudes – EPHE, Humanities and Social Sciences. Jackson, R. & G. Burleigh 2007 ‘The Senuna treasure and shrine at Ashwell (Herts.)’ in: Haeussler & King (eds), vol. 1, 37–54. Jufer, N. & T. Luginbühl 2001 Les dieux gaulois: répertoire des noms des divinités celtiques connus par l’épigraphie, les textes antiques et la toponymie. Paris, Errance. King, A. C. 2005 ‘Animal remains from temples in Roman Britain’, Britannia 36, 329–70. King, A. C. 2007. ‘Romano-Celtic temples in Britain: Gallo-Roman in luence or indigenous development?’, in: Haeussler & King (eds.), vol. 1, 13–8. Koch, J. 2013. Tartessian. Celtic in the Southwest at the Dawn of History. 2nd revised and expanded edition. Aberystwyth, Celtic Studies Publications. Lambot, B. 2006. ‘Religion et habitat. Les fouilles d’Acy-Romance’, Religion et société en Gaule, ed. G. Goudineau, 177–90. Paris, Editions Errance. Lejeune, M. 1981 ‘En marge d’une rigani gauloise’, Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres, 125/1, 29–30. Meid, W. 2003 ‘Keltische Religion im Zeugnis der Sprache.‘ Zeitschrift für Celtische Philologie 53, 20–40 (= Gorrochategui, P. & de Bernardo Stempel, P. 2004, Die Kelten und ihre Religion im Spiegel der Sprache, Vitoria 2004, 175–95). Pechoux, L. 2010 Les sanctuaires de périphérie urbaine en Gaule romaine2010, Montagnac, Éditions Monique Mergoil (AHR-18). Roth Congès, A. 1997 ‘La fortune éphémère de Glanum : du religieux à l’économique (à propos d’un article récent)’, Gallia 54, 157–202. Sims-Williams, P. 1998 ‘Celtomania and Celtoscepticism’, Cambrian Medieval Celtic Studies 36, 1–36. Swift, C. 2002 ‘Celts, Romans and the Coligny calendar’. TRAC 2001. Proceedings of the Eleventh Annual Theoretical Roman Archaeology Conference, Glasgow 2001, ed. M. Carruthers et al., 83–95. Oxford, Oxbow Books. Wooding, J. 2009 ‘Reapproaching the Pagan Celtic Past – Anti-Nativism, Asterisk Reality and the Late-Antiquity Paradigm’, Studia Celtica Fennica 6, 51–74. Zavaroni, A. 2007 On the structure and terminology of the Gaulish calendar, British Archaeological Reports British Series. Zwicker, J. 1934–1936. Fontes historiae religionis celticae. Berlin: de Gruyter. Hae ussle r & King [ 35 ] TA BL E 1 - F.E. R .C .AN . WORKS H OPS Workshop I II Year 1998 1999 Location Vienna, Austria Luxembourg IV 2002 Osnabrück, Germany V 2003 Graz, Austria VI 2005 London, England VII 2006 Cascais, Portugal VIII IX 2007 2008 Gargnano, Italy X 2010 Paris, France XI 2011 Erfurt, Germany XII 2012 Berlin, Germany XIII 2014 Lampeter, Wales III XIV XV 2000 2015 2016 Vitoria/Gasteiz, Pays Basque, Spain Molina, Spain Trier, Germany Lisbon, Portugal Conference Proceedings not published not published J. Gorrochategui & P. de Bernardo-Stempel (eds.), Los Celtas y su religión a través de la epigra ía, Actas del III Workshop F.E.R.C.AN. Vitoria-Gasteiz 2004. R. Wiegels & W. Spickermann (eds.), Keltische Götter im Römischen Reich. Akten des 4. internationalen Workshops “Fontes Epigraphici Religionis Celticae Antiquae“ (F.E.R.C.AN.) vom 4.–6.10.2002 an der Universität Osnabrück. Möhnesee, Bibliopolis. 2005. M. Hainzmann (ed.), Auf den Spuren keltischer Götterverehrung. Akten des 5. F.E.R.C.AN.-Workshop, Graz 9.–12. Oktober 2003, Wien/Vienna, Österreichische Akademie der Wissenschaften Wien (Mitteilungen der Prähistorischen Kommission, volume 64). 2007. R. Haeussler & A. C. King (eds.), Continuity and Innovation in Religion in the Roman West, Portsmouth, Rhode Island (JRA Supplements 67 & 67.2), vol. 1 & vol. 2. 2007 & 2008. J. d’Encarnação (ed.), Divindades indígenas em análise. Divinités pré-romaines - bilan et perspectives d’une recherche. Actas do VII workshop F.E.R.C.AN., Cascais, 2527.05.2006 Cascais. Coimbra/Porto. 2008. A. Sartori (ed.), Dedicanti e cultores nelle religioni celtiche : 8. workshop F.E.R.C.AN., Gargnano del Garda, 9–12 maggio 2007. Milano, Cisalpino. 2008. J. Arenas Esteban (ed.), Celtic Religion across Space and Time. Molina de Aragón & Toledo, 2010. A. Hofeneder & P. de Bernardo Stempel (eds.), Théonymie celtique, cultes, interpretatio / Keltische Theonymie, Kulte, interpretatio. Wien, Verlag der Österrechischen Akademie der Wissenschaften, 2013. W. Spickermann (ed.), Keltische Götternamen als individuelle Option? Celtic Theonyms as Individual Option? Akten des 11. internationalen Workshops „Fontes Epigraphici Religionum Celticarum Antiquarum“ vom 19.–21. Mai 2011 an der Universität Erfurt. Osnabrück (Osnabrücker Reihe zu Altertum und Antike Rezeption, vol. 19). 2013. Published as part of: W. Eck & P. Funke (eds.), Öffentlichkeit – Monument – Text. XIV Congressus Internationalis Epigraphiae Graeca et Latinae, 27.–31. Augusti MMXII. Akten (Corpus Inscriptionum Latinarum, Auctarium, series nova, XIV). Berlin: De Gruyter. This volume. Kelto-römische Gottheiten und ihre Verehrer. Akten des 14. F.E.R.C.AN.-Workshops Trier 12.–14. Oktober 2015 (Pharos 39). Rahden/Westf : VML, Verlag Marie Leidorf GmbH, 2016. Forthcoming.

RELATED PAPERS

RELATED TOPICS

Log In

Celtic Religions in the Roman Period: Personal, Local, and Global - Introduction

Celtic Religions in the Roman Period: Personal, Local, and Global - Introduction

Related Papers

RELATED PAPERS

RELATED TOPICS