The Oak and the Thunder-God Author(s): H. Munro Chadwick Source: The Journal of the Anthropological Institute of Great Britain and Ireland, Vol. 30 (1900), pp. 22-44 Published by: Royal Anthropological Institute of Great Britain and Ireland Stable URL: http://www.jstor.org/stable/2842615 Accessed: 10/12/2010 11:29

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=rai.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Royal Anthropological Institute of Great Britain and Ireland is collaborating with JSTOR to digitize, preserve and extend access to The Journal of the Anthropological Institute of Great Britain and Ireland.

http://www.jstor.org ( 22 )

THE OAK AND THE THUNDER-GOD.

BY H. MUNROCHADWICK, M.A., Fellow of Clare College,Cambridge.

[PRESENTED JANUARY 9TH, 1900.]

INTRODUCTION. THE objectof this paperis to discussthe connectionbetween the sacred oak and the cultof the thunder-god.For this purposeit is necessaryto give an account, first,of the cult of the thunder-godas practisedby the chiefpeoples of Europe, and secondly,of the tree-sanctuarieswhich are foundamong the same peoples. Afterthis we may considerthe evidencefor connectingthe sacred oak with the cult of the thunder-god;and in conclusionan attemptwill be madeto give some explanationof this connection. Since the mythsand cults of the Greeks and Romansare comparativelywell known,it seemedunnecessary in most cases to do morethan give referencesto easily accessibleauthorities. On the otherhand I have thoughtit advisable to illustrateat soniewhatgreater length the less familiarreligions of thenorthern peoples. In the courseof the discussionI hope to show that the thunder-godwas the chief deity of the early inhabitantsof Europe,and furtherthat the temples of late heathentimes were, to a greatextent, developedout of tree-sanctuaries.I think,therefore, it will be admittedthat the subjectis one of considerableimportance for the study of earlyEuropean religion.

CHAPTEI 1:.

?1. ThexThqvtder-Go(i in the North. H. Petersenin his book On-Nordboernes Gudedygikelse og Gudetro i Hedenold, has shownthat in the Northduring the last centuriesof theheathen age Thorwas moreworshipped than any othergod, and that his cutltbears everysign of a remoteantiquity. This subjecthas beenso fullyand ably discussedby Petersen thatit is needlesshere to do morethan brieflyrecapitulate the main arguments whichhe has advanced. Theyare as follows:-(1) Whenevermention is made of figuresof the gods in temples,Thor's figure seems always to take the chiefplace. In no case is it statedthat the centralposition was occupiedby any othergod. In severalcases Thor'sfigure is the onlyone mentioned. (2) In compoundpersonal names /hd- is yastlymore frequent than the nameof any othergod. Thus in the Landn4ina Bo6kTh&or- o-curs in 30 miieln'salnd 21 womenl'sniameis, distributed over about800 H. MUNROCHADWICK.-The Oak flndthe Thunder-Cod. 23

and 300 personsrespectively. On the otherhand Frey-occurs only in twomen's namesand one woman'sname (four persons in all), while 6i5dn-does not occur. The significanceof thesestatistics is shownby thefact that persons who borethe nameof a god were regardedas being underhis special protection.' In place- namesalso Tlhdr-is farmore irequent than the nameof any othergod. (3) When Scandinaviangods are mentioned by foreign writers, either Thor is mentionedalone or he is representedas the chiefof the gods. (4) Especiallysignificant is the title AittgardtsTzqtrr " sanctifierof the earth" whichis applied to Thor in the Edda. Petersenillustrates this by the use of the phrasebur uiki Jisi kuml (or Y.u. Pasi runar) "may Thor sanctifythese mounds" (or " these letters") which occurson two monumentsin Jutland. The same idea is symbolicallyexpressed by thereprcsentation of the hammeron severalinscriptions in Jutlandand Sweden. That thehammer was used at funeralsis madeprobable by the accountof Balder's funeralin Gyjlfaginning49, whereThor consecrates the pyrewith his hammer. A similar usage at weddingsnmay be inferredfrom the story in fThrymskvida. Peterseninfers that the hammerwas used on many otherimportant occasions, especiallyat the openingof the assembly. Hammersused as personaladornments have frequentlybeen found. That theywere used also in publicworship is made probableby Saxo's account (xiii, p. 630) of the large metal hammers(malleos loticales)which Magnus Nielsen carried away froma templeof Jupiteron one of the Swedishislands (about A.D. 1130). As sanctifieraild guardianof the home Thor's figurewas carved upon the &ndvcgisszlur"pillars of the high-seat"(the place of honourreserved for the head of thehouse). The reverenceattached to thesepillars is shownby several storiesdealilng with the colonisationof Iceland. The colonistsbrought their high-seat pillars with them. Thorolfrthrew the high-seatpillars of his templeoverboard on approachinglanid aind accepted the directionwhich they took as an indicationwhere to fixhis newhome.2 Hallsteinn, son of Thorolfr,prayed Thor to send himhigh-seat pillars.3 (5) It is likelyalso thatThor was regardedas the patronof theassembly. In Iceland,Thursday was the day on whichthe Al-ingi (generalassembly) was opened. This was also the case withthe Gulapingin Norwayand withmost of the districtassemblies in the Danish islands. (6) Evidencefor the cult of Thorin timeof war is givenby the accountof Styrbiorn'scampaign against Eirekr, king of Sweden,and by Dudo in his accountof the Normans. Snchare, according to Petersen,the mainfeatures of the cult of Thor. The importanceof the cult maybe estimatedby the fact that it embracesevery side of publicand privatelife, whereas the cult of Othinis concernedchiefly with the militaryside. LastlyThor seems to embodythe ideal of the nationalcharacter; he is representedas fearless,impetuous, but benevolenttowards men. Othin on the otherhand is distinguishedrather by shrewdnessand cunning.

I Cf. EyrbyggiaSaga, c. 7, 11, and the passage from Grwnl.Anndll. appended to Vigfusson's edition of the Saga. 2 Eyrbyggia,s. 4. a LandndmaBOk, i, 23. 24 H. MUNROCIIADWiOK.- 1ThC Oak and the Ikunder-God.

It is unfortunatethat no figureshave suirvivedwhich may be identifiedwith certaintyas representationsof Thor. Yet therecan be little doubt howhe was depicted,for the, Sagas containseveral storiesof his appearingto his friendsor enemiesupon earth.' He is representedas a man of largeand powerfulbuild, in theprilmle of life,and havinga red beard. The equipmentof thegod as depicted in themythological poems is renmarkable,especially from the negativeside. His weaponis almostalways the hammer.2He is never representedas possessing spear,sword, shield, helmet, or coat of mail. Againin travellillghe eithergoes on footor drivesin a carriagedrawn by goats. A horse is neverascribed to him. In Grm12nnismll29, immnediately before the enumerationof the horseswhich the godsride when they comneto do justice underYggdrasill's Ash, it is statedthat Thorhas to wade throughseveral streams on his way thither.The horsesof Othin, Freyr,Heimdallr and Balderare mentionedalso elsewhere. The antiquityof the representationof Thormay be estimatedby the absenceof the horseand of all the ordinaryweapons of war. His equipmentindeed resembles that of a hero of the StoneAgte rather than of any sutbsequentperiod. Bothfronm the shape of the hammeras depicted onl Rlunic stones,etc., and fromthe fact that the word (O.Norae Itarnarr)also means "rock" and is perhapsrelated to O.Bulg. kamy "stone" and kinldredwords, it appears n1otunlikely that Thor's hammerwas originiallya stone implement. Humansacrifices in connectionwith thiscult are seldommettioned. They were known,lhowever, both among the Norlmans(Dn)do, i, 1) and in Iceland (Eyrbyggia,s. 10), thoughin the lattercase the victimiiswere probably condemned criminals.4It is solmiewhatremarkable that in both these cases death seems to have beeninflicted by fellingwith a woodeninstrumiiient. Elsewhere sacrifices of horsesand oxen5and offeringsof bread and meat6are miientioned.Besides the sacrificesthere is one otherpoint which deservesmiientioln. In Icelandic temples it seems to have beenthe customto keep a sacred firein the aflh?'s(cj: p. 28) whichwas neverallowed to die ouit. Since the templedescribed in Kialnesingas.2, wherethis notice occurs, was primarilya Thor-temple,it is likely that thislholy firewas connectedwith the cult of Thor. It has been shownby Petersen(see above) that the blessingsconferred by Thorapply to all departmentsof human life. His connectionwith weatherand natural phenomenais surprisinglylittle evidenlcedin Norwegian-Icelandic literature. This is no doubt due in part to the disuse of the old word for "thuinder."In Denmarkanid Sweden on the otherhand, where the wordstorden and a'ska (toraka)are preserved,the conniectionseems to lhavebeen moreclearly kept. But it is ratheras the protectorof the humanrace against trolls (the

1 Cf.especially Fornmanna Sgyur, ii, 182. 2 Saxo, iii,p. 118,provides him with a club. 3 Cf.Kialnesingas. 2. For a similarcustom among the Gatils of. Ccesar, RG., vi, 16. 5 Plateiarbok,ii, 184. ( Olafss. helga,33ff. H. MUNROCHADWICI.-TShe Oak-acntl the Thunder-God. 25

spiritsof thedesert), as thegranter of land,the guardianof the laws and customs of civilisedlife that Thor is best known. In connectionwith Petersen's suppositionthat Thor was regardedas the patronof the assemblyit is perhaps worthrecalling Grimnismlt 29, whereThor is representedas proceedingto do iusticeurnder Yggdrasill's Ash. None of the othlergods are miienitionedby name in thispassage. In conclusion a few words must be said in reg-ardto Thor's family relationships.He is called the son of Othin,but this canihardly be an original feature,for his cult appearsto be muclholder than thatof Othin. His motheris called Fi6rgynlor I6rd (" earth"). His wife is ,2his sons Mo6di (" the courageous") and Magni (" the mighty"), his daughterpru'dr.3 The lateness of the last fournames is shown by theirallegorical character. The lnatureof Thor'srelationship to the human commulnityis well illustrateclby the choice of Sif as thename of his wife. Lastly it is worthobserving, that no royal or noble familyseems to have tracedits genealoayto Thor.

? 2. TlheThunder-God of theContinental Germans. Among,the continentalGermans and in England veryfew traces of this cult havesurvived. In the " Old Saxon"' renunciationformula l'Thnnaer is mentioned toaetherwith TWIoden and Sacxnot.Aaain the inscriptionon the Nordendorfbrooch, the meaningof whichis not altogetherclear, ends with the wordswigi Ponar, whichcorresponds exactly to the formulaemployed on certainimoniumental stones in Jutland(cf. p. 23). The earliestcertain evidence for the existenceof the cult is the phrase O.H.G. doncarestag, O.Enal. Punr}esdceg as a translationof Lat. dieslouis.4 It is worthobserving, that amiiong the in the sevenithcentury St. Eligius (t 659) found it necessaryto forbidthe celebrationof Thursday festivities.5With this may perhapsbe comnparedthe fact thatin manyparts of GermanyThursday has been the favouriteday for weddinigsdown to modern times. The Germanthunder-god is sometimesmentioned under his Latin namie, e.g.in the IndiculusSuperstitionutm and in Bhonifacius,Epist. 25, wheremention is madeof a priestwho " sacrificedto lupiterand ate the sacrificialmeat."6 It is uncertainhow far the attributesby whichThor was disting,uishedin the North were applied to the thunder-godon the continent. In the Frisian

1 Identicalwith Goth. fcirguni "mountain," O.Engl. firgen- " forest,"etc. 2 The samieword is foundin the otherGermanic languages (Goth. sibja, O.Engl. sib, O.Sax. sibbia,O.H.G. sippa),in the senseof "relationship"(in O.Engl. and 0.1EG. also "friendship," " peace"). In Old Norseonly the pl. sifiar" relationship,"occutrs. 3 Identicalwith O.Engl. pry} " strength." 4The namesof the days of the week were borrowedprobably between the secondand fourthcenturies. 6 Grimnm,Deutsche Mythologie4, iii, 402. 6 It is somewhatremarkable that the Germaniclupiter is liot mentionedby the earlier Latin writers. The only god besides Mercuriusand Mars to whomnTacitus refersat all frequentlyis "Hercules,"and it has beengenerally assumed that tunderthis name he meant the thunder-god.This, however, does not seem to me to havebeen satisfactorily established. 26 H. MUNROCHADWICK.-The Oak and theThunder-God.

storyof the twelveAsegen,' if the implementborne by the god is an axe,2it is verylikely that it is the thunder-godwho is meant. If so thiswould be valuable evidencefor provingy the connectionof the thunder-g(odwith the laws. Another possiblereference to the axe of the thunder-godoccurs in the Old Englishdialogue of Salomonand Saturn(ed. Kemble,p. 148).3

? 3. fI'e KelticThunder-God. Very little is known about the gods of the ancient Kelts. There was, however,a god named Taran, who is mentionedby Lucan (I. 446) and whose name appears also in inscriptions as Tanarus, TLaranucus, Taranuenus.4 Lucan says that he was honouredwith human sacrifices. Now the form Tanarus correspondsexactly to *punuraz,the Germanicname for thunder and the thunder- god,while Taran- differstherefrom only by metathesisof -n-and -r-. The latter maybe identicalalso with Mod. Welsh taran " thunder." It is knownthat the Kelts had a god " Jupiter"; Cmesar(B.G., vi. 17) states that,like othernations, theyregarded him as the rulerof the gods. Since the Latin Iupiteris likewise the thunder-god,there can be littledoubt that Taran is thegod meantby Cmesar. Thisconclusion is furtherconfirmed by the dedicatoryinscription 1; 0. M. TLanaro foundat Chesterand datingfrom A.D. 154.5 It is likely enoughthat the god correspondedto some extentin attributes,as well as in name,to the Germanic thunder-god,for there seems to have been much similaritybetween the two nations. But the evidenceis not sufficientto elnableus to arriveat any certain conclusions. ? 4. The Baltic Thuncer-G'ocl. The evidencefor the cult of the thunder-godamongst the ancientPrussians is muchmore extensive, but unfortunatelyit is late and notfree from suspicion. Grunaugives the god's name as PerkunoGand says that togetherwith the gods Patollo and Potrimpohe was believedto inhabitthe sacredoak at Romove. He was supposedto communewith the prieststhere in thunderstorms.In thisway thefundamental laws of the nationwere believedto have beenimparted by him. In his honoura perpetualfire was kept burningunder the oak. The priestwho allowedthis to die out was condemnedto death. According,to Grunau(Tr. II, cap. v, ? 1) Perkuno'slikeness, as depictedon King Widowuto'sbanner, was that of a middle-agedman with black beardand wrathfulexpression of countenance; I v. Richthofen,Friesische Rechtsquellen,p. 439f. 2 The reading is uncertain. 3 There is no direct referenceto the god in this pasage, but cf. Saxo, xiii, p. 630. 4 Taranucus in a Dalmatian inscription (C.I.L. 3, 2804); Taranucnus in two Rhenish inscriptions(Brambach, C.I.Rh. 1589, 1812); on these names see Much, Festgabe f. Heinzel, p. 2227f. Cf. also Taranin louen in a scholion to the above passage of Lucan (Usener). Identical with Lith. perkAnas "thunder," which is still used personally in folk-songs; cf. also Michov in Grynaeus, Novus Orbis Terrarum ac Ifslularum Veteribu8Incognitarum, Basel 1537, p. 519, who says of the Samagitti (in East Prussia): clis su2is falsis,prwcipue deo lingua eorumappellato Perkuno, id est tonitru,adfocos quisque S80os offerebatlibamina. H. MUNiROCHADWICK.-The Oak.cad thbeTAunder-God. 27

his head was crowinedwith flames. Among,the holysymbols of the Lithuanians, accordingto YEneas Sylvius,was a huge iron hammer. From the place-name Heiligenbeilit is likelythat similar symbols were uised amonlg the Prussians. At the presenttime it is customaryto regardthe authoritiesfor the ancient Prussianreligion, especially Grunau, with the greatestscepticism. Soimewriters have evengone so faras to doubtthe existenceof a god Perkuno. This,however, is certainlyunjustifiable.' What especiallymakes for the credibilityof Grunau's accountin the main,distorted and embellishedthough it is withoutdoubt, is the fact that thereis scarcelyone of the religiousobservances mentioned by himfor whicha parallelcan not be founidin someother European people, generally at a veryearly period of history.2In manycases theseforeign customs can not have been known to Grunau. ThleNorthern cult of Thor offersseveral points of resemblanceto Grunau'saccount of Perkuno. It has been shownabove thatthere is some reasonfor regarding Thor as the god of the Assem-bly.This goes far towardsconfirming the conceptionof Perkuno as the law-giver. Again the perpetualfire in honourof Perkuno may be comparedwith the perpetualfire whichwas kept burningin Thor'stemples in Iceland. Lastlythe portraitureof Perkunoon the bannerstrongly resembles the Northernportraiture of Thor. The banneris Ino doubt fabulous,but the picturemay have been drawnlfrom figuresof the god such as the one whichis statedto have stoodin the oak.3

? 5. The Slavonic Thunder-God. In the treatiesmentioned by Nestor(Ch7ron. 21, 36,etc.) the Varangianscall to witnesstheir god Perun. It is very likely that the Northerngod Thor is meant,but he would seemto have been identifiedwith a native god and to have adoptedthe Slavonic name.4 That the cult of the thunder-godwas nothingnew amongthe Slavs is clear fromProcopius' statement (, iii, 14) that "they considerone god, the creatorof the lightning,to be the onlylord of all things." ? 6. The Thtnder-Godamongst the otherpeoples of Europe. It is unnecessaryhere to discuss the cult of the thunder-godamong the ancientGreeks and Romans. That the Greek god Zeus bore this characteris

' Cf. Zeuss, Die Deutschenund die Nachbarstdmme,p. 41. It should,moreover, be rememberedthat the latestof the writerswhom I have quoted,Grunau and Michov,wrote withina centuryof the death of Vitoldus(Vitovt), in whose reign(1392-1430) the Baltic tribeswere still mainlyheathen. .iEneasSylvius (Pius II, Pope 1458-1464)is practicallya contemporaryauithority. 2 I hopethat in severalcases this parallelism will be made clear in thefollowing pages. 3 It is possible,though hardly likely, that the religionof the Prussianswas influenced by thatof the Northin late heathentimes. Gothicinfluence at a verymuch earlier period is moreprobable. 4 The name is identicalwith Russ., tech. perun"thunderbolt," and probablyalso with Lith. perk4nas,though the loss of the -k-is still unexplained Much (Festgabef. Heinzel, p. 212ff)-holds that bothperun and perkatnasare loan-wordsfrom Germanic, but his theory seemsto me to be besetwith overwhelming difficulties. 28 H. MuNrtoCHADWICKi.-Thl Oak and the Tlunder-God.

clear frolmithe epithetsapplied to Ihimsuch as Trp=rKe'pavvoo9,ap7yKepavvog v* 3pe/T179, ept/3p.eTr7,o-rEpo07rryepera. He was also a god of therain. In this connectionit is wortlhnoting, that a raini-charmnwas in timeof droughtpractised by the priestof Zeus oniMount Lykaios in Arcadia. The ceremonyconsisted in dippingan oak twigin the sacred strealmi(cf. Paus. viii,38, 3). With this may be comnparedthe Roman)aqguwlicium which took place at the templeof Iupiter on the Capitol.' The connectioniof the god Iupiterwith the thunderand with the phenomeniaof the weatherin general is so well knownthat it will be sufficientherde to referto Preller,Rintische Mythologie, p. 190ff:

? 7. Conclussion. The foregoingbrief sunmmaryis enough to show that the cutltof the thunder-godwas in earlytilmles common to mostof the Indo-Germaniicspeaking peoplesof Europe.- The Greekand Romanthunder-gods, as mightreasonably be expectedfrom the identityof theirnames ZevI5 (Voc. ZeD ra'Tep,Hom.)-Iut-piter (Iquppiter),seem to bave lhad commrionfeatures evenl apart fromithe influence whichwas exercisedby the one people upon the otherin historicaltillmes. It mayreasonably be supposedalso that such was the case withthe Keltic I'anaros (Tarctnos) anidthe Germanicpunuraz, where there is the sanmeidentity of name. Lastly,it has been shownthat the Prussialncult had severalpoints of resemblance to thatof the northernGermans. Here also a simriilarityof names is notwanting, forPerkenas is closelyrelated to PiUhgyn,the nameof Thor'slmlother.

CIIAPTERIL

? 1. Th1e Gervanic Tree-Sanchtary. The Northierntemiiple in late hleatlhentimiies seermis to have beenishaped miiuchl like a churchwith an apse-shapedbuildilng (afh'is) at the east enid. The afhsts, whiclhin Iceland seems not to have been openito the rest of the building,was the sanctuaryproper and containedthe figuresof the gods,the altarwith the oath-ringand blood-kettleand the perpetualfire. Petersen,however (op. cit. hap.2), believesthat therewas anotherform in use, namelythe round temple, of whichhe thinksthe roundchurches, which occur in all Teutoniclands, are a relic. He believesthat this is the moreoriginal form and thatit is a developmentof the grove-sanctuary.In the moreusual type of templethe afhqswas the original sanctuary,while the long rectangularbuilding (langhAs) was a lateraddition and due to the exigenciesof the sacrificialfeast.2 Temples were known amolngthe other heathen Gernmanicnations in the earlymiddle ages. In the sixth centurya Frankishtemple was destroyedby

1 Tertullian,Ap. 40. 2 It is possible,of course,tha thieform of t,h temnplewas iinfltuencedby thatt of Christian churchesin the BritishIsles. H. MUNROCHADWICK.-IVliC Oak t1and the Thztnder-God. 29

Radegunid,wife of Chlotar.1 Gregoryof Tours' describesa templeat Cologne whiclhcontained figures of the gods and in which sacrificialfeasts were held. In Auaustine'stimne the Englishhad templeswhich were capable of beingturned into churches.3The Frisianisand Old also seem to have had temples, but the notices are not always clear. The word fanzGm, by which these sanctuaries are usuallydenoted, is also used of the " ,"which was an uprightpillar. In othercases again it is clear that the Germanicilnvaders had appropriated .Romanitemples to the use of theirown gods. The trueGermaniic temples, except in Iceland,seem to have been usually,if notalways, made of wood. In earliertimiies the evidencefor the existenceof Germanictemples is very meagre. The clearestcase is Tacitus,Ann. i, 51,where it is statedthat Germi'anicus " rasedto theground the temple which they called the temiiple of Tamfana and which was most frequentedby those tribes." In Germ.40 miention.is made of the "templum" of ,but it is not unlikelythat the word is looselyuse(d for "sanetuary" and refersto theholy grove miientioned just before. Accordingto Gern. 9 the Germans"cdeem it to be inconisistentwith the majestyof the gods to confinethem within walls or to representthem arter any similitudeof a humanface; theydedicate groves and woodsand call by the niame of gods that invisiblething which theysee onlywith the eye of faith." Sacred grovesare mentionedby Tacitusin several otlherplaces. Accordinigto Germ.39, the Semnoneshad a wood of immemorialantiquity anid holiness; accordingto Germ.43, the Nahanaruali had a groveof ancientsanctity. The sanctuaryof Nerthus(Germ. 40) was an island grove. 'forces assemble in a wood sacredto Hercules(Ann. ii, 12); Civilisbrings his armytogether in a sacredwood (Hist.ii, 14). Thesesacred groves contained altars (Ann. i, 61), and eftgicset signa quaedamwhich the priestscarried into battle(Gerin. 7). Witlhthe last passage miaybe compared list.iv, 22,from which it appearsthat the etigieswere figures of wildbeasts. The accountsof later writersfully agree witlhTacitus' evidence. Claudian (Cons.Stil. i, 288) speaks of "groves grimIwith ancientreligious rites and oaks resemiiblinga barbaric divinity" wlich he hopedthe RIomanaxes would fell. In the TranslatioS. Alexandri(Mon. Germ12.ii, 676) iG is stated that the Saxonis worshippedtrees and streams. Similar evidenceis giveniby the occurrenceof place-namessuclh as Heiligenloh,Heiligeniforst, Heiligenholtz. In the North place-namesending in -lund probablydenote the formerpresence of grove- sanctuaries. There was a holygrove beside the Swedishtemple at Upsala, one evergreentree being regardedas especiallysacred. The legendarysanctuary at Glaesisvdllr(Hervarar Saga 1; Forn. S&g. i, 411) was probably connected with a holygrove; Glcesirhere is, perhaps,identical with Glacsir, the nam,e of the tree in Valholl(Othin's dwelling-place). Thereis an obviousconnection between these sacred groves and the " guardian IActa Bened,s. i, p. 327. 2 VitacPatrutm, 6. Bede,B lEist. Eccl., i, 30. 30 H. MUNROCHADWICK.-Te Oak and the,Thunder-God.

tree" (Vardtrad)which is foundstanding beside the homesteadin manydistricts of Sweden." Mannhardtdescribes how formerlyevery house in the sailors' quarter of Copenhagenpossessed an elder tree which was tended with the greatestreverence and regardedas the dwelling-placeof the guardianspirit of the house. In the Northernmytholoay Valhbll has beside it a sacred groveor tree (Lceratr,). I have tried elsewhere2to show that the evergreentree at Upsala was regardedas theVardtrad of the Swedishnation and thattlle conception of the " world tree,"Yggdrasill's Ash, mrlayhave arisen fromthe idea of a Vardtradof the divinecommunity. We may now return to the discussionof the Northern temple. The "langhus" seems to have closely resembledthe hall of an ordinarylarge dwellinghouse. Like this it containedthe " high-seatpillars " which stood beside the seat of the ownerof the temple. Indeed therecan be little doubt thatthe langhus is copied directlyfrom the hall and thatoriginally it formedno part of the templeproper. Originallyit would seemto have beenthe residence of thepriest or chiefwho presided over the templeand providedthe feasts. Petersen's suggestion that the " afhu's" (the sanctuaryproper) is a development of the holy grove is rendered probable by the following considerations:-(l) The temnpleappears to have been of comparativelylate originamong the Germanicpeoples. The passage quoted above fromTacitus (Germ.9) practicallyamounts to a statemelntthat the Germanshad no temples, thoughin one or two instancestemples seem to be actuallymentioned by him. (2) The Old Englishword hearg, herg, which is used to translatesacellnm, lupercal, simulacrrum,fanum, templum, is apparentlyalso used for"grove" and is identical withO.H.G. harucwhich is used to translatelucus, nemus, fanum, ara.3 (3) The sacredtree or groveis sometimesfound standing beside the temple,as at Upsala. The same phenomenonis founldarnong other European peoples, as in thecase of the oak at Stettindestroyed by BishopOtto (Qf p. 33), and the pine destroyedby St. Martin(cf. p. 34). In everycase the tree seems to have been regardedwith greaterreverence than the temple. (4) In other European countriescertain templesare distinctlystated to have been erectedon the site of a holytree or grove. Such was the case withthe templeat Dodona and,according to tradition, with the templeof Iupiter Capitolinusat Rome. From these considerationsit seems to me not unlikelythat if the langh(s is a developmentof the chief's dwelling,the afhsismay bear someclose relationshipto the sacred groveor tree (Vardtrad)which originally stood beside the dwelling. In conclusionit is worthmentioning that the sacredgroves of the ancient Germansseem to have beenused also as places of assembly,the meetingof the Thingbeing no doubtclosely connected with religious festivals. Passageshave been quoted above(p. 29) showingthat the Germanarmies assembled in sacredgroves 1 Mannhardt,Baumkultus, s. 51ff. 2 Cttltof Othin,p. 75ff. *3 Withthis may be comparedO.H.G. paro,which is usedto translateboth lucia and ara, and is identicalwith O.Enol. bearu " grove." H. MUNRO CHADWICK.-The Oak-and theThitnder-God. 31

duringtheir wars with the iRomans. It is probablethat at the nationalassemblies in the wood of the Semnones(Germr. 39) the businesstransacted was politicalas well as sacral. Certainlythe great assemblies at theUpsala grove-sanctuarymet for morethan merely religious purposes. The usual namefor the place of assemblyin theNorth is Pingv1llrwhich must originally have meant" wood of the assembly," thoughvollr has come to mean" field" in the North.' It is iD a forest-clearing thatthe godsmeet to decidethe fateof Starkadrin GautreksScaga 7.- According to a mythologicalpoem (Grn,rnn. 30) the gods nmeetevery day to dispensejustice underYggdrasill's Ash. ? 2. The Baltic Tree-Sanctuary. The onlyreference of which I knowto the existenceof templesamong, the Baltic tribesis a passagein Michov2 whichdescribes how Vladislaus extinguished the sacred fire at Vilna and destroyedthe ternplurnet ara at which the Lithuaniansoffered their sacrifices. That thiswas a real templeis made probable by the factthat shortly afterwards he speaksof the turrisin whiclhthe sacred fire was kept. On the otherhand thereis no otherEuropean peoplewith whom the sacred grove ocPupiesso prominenita position. ?Eneas Sylvius (Hist. de Eur. xxvi) describingthe missionaryjourneys of Jeromeof Prag among the Lithuanians, statesthat he came to a peoplewho worshippedwocds dedicatedto devils. There was one wood sacred above all others,and in its centrean oak of greatag,e which was especiallyregrarded as the abode of theirgod. As Jeromecontinued to cut thesetrees down, a greatcrowd of womenbetook themselves with lamentations to Vitoldus'complaining, that " theirsacred grovehad been cut downand the house oftheir god takenaway; in it theyhad been wontto solicitthe divinefavour anid fromit theyhad obtainiedraini and sunshine; now theydid not knowwhere to seek the god whose habitationthey had takenaway." Michov(I.e. p. 518) states thatthe Lithuanians "deemed woods and grovesto be holyand regardedthem as the habitationsof the gods." In a subsequeentpassage he relates that the Samagitticonsidered even the birdsand beasts which enteredthe sacred woods to be holy. Theybelieved that if anyoneinijured these woods or anythin(gin them, thc devilswould make his handsor feetto growcrooked. Accordingto Erasmiius Stella (de BorussiaeAntiq2ttatibus, ii4), the Prussianssaid that the gods dweltin grovesand woods; heresacrifices were to be offeredto them; fromhence sunshine and rainiwere to be obtained. " Theysaid thatthe gods inihabitedthe finesttrees, such as oaks; fromthese trees enquirersheard repliesgiven to them; therefore theydid liot cut downtrees of thiskind but tendedthem religiously as the houses of theirdeities. Theytreated the alderand severalother trees in the same way." 1 The wolrdis identical with O.Ellgl., 0.1-H.G.wald " forest." A similar change of meaning (fromi" grove" to " nleadow ") is seen in O.Engi. lf I. 2 Grynaeus,SoVms Orbis,etc., Basel, 1537, p. 518f. 3 Prince of Litlhuaniia,see p. 27, footnote. 4 Grynaeus, Yovus Orbis,ete., Basel, 1537, p. 581. 32 H. MUNROCHADWICK.-Thte Oak aGidthe Thunder-God.

Accordingto Helmold(Chrqon. Slav. i, 1), Germanswere, in his day,admitted to all privilegesamong the Pruzi (Prussians)except the rightof access to grovesanld springs;they thought that these would,be polluted by the preseniceof Christianis. A similarstatement is made by Stella (I.C.) who adds that atonementfor the intrusionof strangerscould onlybe made by the sacrificeof a humanvictim. Among the Letts a custom is foundsinmilar to that coninectedwith the Vardtradin the North. Accordingto Mannhardt(Baumnoultus, p. 52ff)it was usual even up to the presentcentury to findbeside the homesteada small grove whichwas regardedas the dwelling-placeof theMalfCts kuttqngs (" lord of thehome ") and honouredwith small offerings. Amongthe Prussiansthe nearestapproach to a tempieseems to have been the holyoak at Romove(Rickoiot), the chiefsanctuary of the Prussians.' This was an evergreentree with foliageso thick that,according to Grunau,no rainl couldpenietrate it. In the stemstood busts of the threechief gods, and in frontof each of these was placed the emiblemof his cult, Perkuno'semblemii being the perpetualfire. PRouindthe oak were huingfinie silk curtainswithin which nloone was' allowed to enterexcept the chief priests; the curtainsmight, however, be partiallywithdrawn to enablevisitors to see the busts. The priestslived in tents round the oak; accordingto one authority2the kirwcvito3or high-priestlived within the curtaini. In this oak Perkunowas believed to communewith the kirwaito,who was knownas "God's mouth." The otherpriests and priestesses (waidlottent)were old widowersand widows,livilng in celibacy,and had to miiake knownto the people the commialndsof the kirwaito. In the neighbourhoodof the oak the Prussians held their religious,and no doubt also their political, assemblies. Michov(I.e.) states that on the firstof Octoberthe Samaagitti4held a great festivalin theirsacred woodsat whiclhthe whole populationcame togetherand sacrificedto theirgods, especially Perkunus. He states,miioreover, that each fanmily had in thesewoods a hearthat wlich theyburlnt their (lead,and seats miacieof cork,oni wlichl were set foodanid drinik for tlle souls of the departed.

? 3. ThezS57aion ic lTee-Sanctutlary.

Holy trees and groves are also among the Slavs. Tlhietmarof Merseburg(Mon. Germn. v, 812) statesfouniid that Riedegostwas comnpletelysurrounded by a forest,which was regyardedwith venerationalnd never touched by the inhabitants. In a subsequentpassage (I.e. p. 816) he relateshow in the year 1008

Grunau, PreussischeChronik, Tr. ii, cap. v, s. 2; iii, cap. i, s. 2; cap. iv, S. 1. "Piisca Antiquoriim Pruteiioruin Religio " in Respubliccasive Status Regni Poloniae, etc., Leyden, Elzevir, 1642, p. 297,ff). 3 This word seemiisto be i(lentical with Litli. kriwditison whichl Schleicher says: "jezt uinbekant. . . . es iilusz einle holhe,wUlrde bezeichneii." 4 FromiGryiaeus op. cit.,p. f524,it is likely that this passage refers,in part at least, to the assemlbliesat Romove. H. MUNROCHADWICK.-Tlhe OaZo and theThbunder-God. 33

Wigbertdestroyed a grovecalled Zutibure,' which was honouredby the inhabitants in all respectsas a god,and whichhad never been injuredsince the begiinningof time. In the year 1124 Bishop Otto of Bamberg destroyedthree Slavonic templesat Stettin.2 He then preparedto cut down an oak whichwas regarded withgreat veneration by the inhabitants,as theybelieved it to be the dwelling place of a deity. Beneath the oak was a spring. The people,thnough they had allowedhim to destroythe temples,besought him earnestly to sparethe tree. The Slavs,however, as has alreadybeen mentioned,also possessedtemples. One of the templesdestroyed by Ottocontained a three-headedfigure (Triglctus). A templeat Riedegostis mentionedby Thietmarin thepassage to whichreference has been made above. Saxo (xiv, 822ff,841ff) gives a full accountof certain Slavonic templesin Riigenwhich were destroyed by King Waldemarin the year 1169. One of these templesconsisted simply of purple hangings,the roofbeing suipportedby pillars. It containedan oakenfigure of iugieuituiswith seven faces. Anothertemple had an outer enclosureof walls and an ilnnerone consistingrof hangings,the roofhere also being supportedby pillars. In the templestood a woodenfigure of Suantouituswith fourheads. The priestalone was allowed to enter withinthe curtainsand was not allowed to take breath there. Another templecontained a figurewith five heads. These Slavonic templesbear a curious resemblanceto the Prussian tree- sanctuaryat Romove. Boetticher3has made it probablethat amongthe Greeks the earliestfigures of the godswere made out of tree-stems,especially from the stem of that tree which was sacred to the god. The statuLewas in fact a developmentof the natural tree. If a simiiardevelopment may be assumedfor tlle Slavonic temple,the latternmist come from a formlvery similar to the Prussian tree-sanctuary.An oakenimage in a templewhich had curtainsfor walls mliayvery well come froma living oak surroundedwith curtains. Possibly even the multiplicityof heads in the Slavonic idols nmaybe due to the custom,which is foundat Romove,of insertingbusts in the trunkof the tree. These arguments seemto me to makeit probablethat the sanctuaryat Romovefaithfully preserved a typewhich was oncecommoni to the Slavonicand Baltic peoples. The veryclose relationshipsubsisting between these peoples is shownby the resemiblancebetween theirlanguages; for,so faras phonologyis concerned,there can have been little differencebetween them at the beginningof the Christialnera. ? 4. The Keltic Tree-Sanctuary. In spite of the paucityof oulrinformation concerning the religionof the ancientKelts, it is quite clearthat they had sacredgroves alnd trees. Pliny(N. H. xvi,249) statesthat the Gauls considernothing more holythan the mistletoeand the tree on which it grows,provided that this is an oak. He tlhelng,oes on to

" l Apparentlyfor Svetsbor holy forest"; cf. Ruhss.bor' "pi nie-forest.' 2 Msn. Germ.Scipt., xii, 794. 3 Der BaumkultusderHellener ue. Rmoer, p. 21,5ff. VO-L. XXX (N.S. III). D 34 H. MUNROCHADWICK.-The Oak,and theThunder-God.

describethe ceremoniesconnected with the cuttinigof the mistletoe,adding that theynever practise any of theirreligious rites without oak-leaves. In conclusion he proposesan etymologyof Druidae fromGreek 8pi39 " oak."' Lucan {i, 453) says thatthe Druidsinhabit deep grovesamid secludedwoods. In anotherpassage (iii, ;399ff)he describesa g'rove-sanctuaryin the neighbourhoodof Marseilles. In Britainalso the Druids had spcredgroves; Tacitus (Ann. xiv, 29) relates how SuetoniusPaulinus destroyed the Druids'groves in Anglesey. ThoughDruidism was put downby the Romans,tree-worship long continuedin Gaul. Sulpicius Severus(Vita S. Martini,c, 10) statesthat St. Martindestroyed an ancienttenmple withoutmeeting with any oppositionfrom the inhabitants,but on his subsequently proceedingto cut down a pine-treewhich was consecratedto a devil in the neighbourhood,he at once encounteredresistance. It is worthnoting that the Gaulishword for temple, nemneton, seems to have originallymeant " grove." ? 5. The Greek-and Roman Tliree-Sanctuagrtie3. For the tree-worshipof thesenations it will be sufficientto referto Boetticher (op. cit.). Here it need onlybe observedthat in classicaltimes the tree-sanctuary had in mostplaces been either wholly or partiallydisplaced by the temple.

CHAPTERIII. ? 1. The Associationbetween the Thunder-Godand the Oak. In the precedingchapter evidence has been given to show that amongthe Baltic tribes(1) the cult of grovesand trees was exceedinglyprevalent; (2) the explanationof this cult given by the people themselveswas that these groves and trees were the habitationsof the gods; (3) the tree most reverencedwas the oak. Is thereany evidenceto showthat this tree-cultwas connectedwith one god moreclosely than the rest? In mostof the noticesno particulargod is specified, whilein Grunau'saccount the oak at PRomovecontains the busts of threegods. Yet thereis reasonto believethat it is Perkuno,the thunder-god,with whom the reverencefor the oak wasoriginally connected. The thunder-godis essentiallya god of theweather ;2 it is rainand sunshinewhich the Lithuanianshope to obtainfromi theirtrees. Again thoughthree gods are mentiolnedin connectionwith the oak at iRomove,yet it is alwaysPerkuno who appears individually. It is he who speaks withthe high-priestin the oak. Furtherit is statedthat Perkuno's emblem was a sacred fire of oak-woodwhich was kept up perpetually. Lastly Hirt (Idg. ForschungenI, 479ff)has made it probablethat the word (Lith.) perkAnasis

I In spite of what has been said to the contraryit seems to me probable that Kelt. druid. is a derivative of a Kelt. stem draz-"oak," though there is no evidence for preciselythis formin Keltic. The derivation is especially favoured by the Welsh re-formatederwydd "druid" fromiderw "oak." Cf. also the Drui-nemeton(" oak-grove" ?), where the Galatian council met (Strabo, xii, 5, 1.) 2 Accordinigto Adam of Bremen, iv, 26, "Thor presides over thunderand lightning,winds and showers,sunshine and the fruitsof the earth," H. MUNROCHADW1CK.-The Ocakanid the Thunder-God. 35

relatedto Lat.quhercs (Jd1g.perkuus) " oak." Consequentlythe word can originally have meantnothing else than" oaken,"and musthave beenan epithet: " the god of" (or " in ") " the oak." Among the ancient Germans also a similar associationmay be traced. Bonifacius'is said to have destroyedat Geismarin Hesse a huge treewhich the inlhabitantscalled " Jupiter'soak."2 That grovessacred to the thunder-godwere knownamong the ancientEnglish is shown by such place-namesas Junres-ljth which must originallyhave nieant "grove of thulnder." In the North the association seems to be forgotten,yet an importanttrace of it is preserved in the wordFi6rgyn, the nameof Thor'smother. This word is closelyrelated to Lith.Perk(nWas, and like it,in all probability,a derivativeof Idg. *perkuus," oak." In regardto the Kelts it has been shownabove :-(1) that the cult of the thunder-godseems to have beenian importantelement in theirreligion; (2) that venerationfor grovesand trees was prevalentamongst them; (3) thatthe oak was apparentlythe tree most reverenced. The connectionbetween the cult of the thunder-godand the reverencefor the oak is shownby Maximus Tyrius,viii, 8:- "The Kelts worshipZeus; the Keltic image ofZeus is a loftyoak." Amongthe Romansalso the oak was associatedwith the cult of the thunder- god (lIupiter). Several writers3give lists of treeswhich were sacred to various gods,and in theseit is alwaysthe oak whichis associatedwith Jupiter.4 Some of thesepassages may not perhapsbe acceptedas unimpeachableevidence for genuine Romantradition, since the suspicionof Greekinfluence is possible. But thereis otherevidence of a less questionablecharacter. The " civic crown" of oak-leaves whichwas givento a soldierwho had saved the lifeof a comradein war,and which was regardedas one of the highestdistinctions which a Romancould gain, seems to have been associatedwith the cult of Jupiter.5The templeof Jupiter Feretrius on the Capitolstood on the site of a sacredoak, to which,according to Livy (I. 10), Romulusbrought the spolitaopima whichhe had gained by killingthe king of Caenina. That the oak was associatedwith the cult of the Greek thunder-godZeuis is clear fromseveral passages relatingto the oak-sanctuaryat Dodona (see below). The associationwas not peculiarto Dodona,as it was foundalso at the sanctuary of Zeus on MountLykaios. For the rest it will be sufficienthere to referto a scholionon Aristophanes,Birds 480 whichstates that " the oak is sacredto Zeus." The associationtherefore between the oak and thethunder-god is foundamong thePrussians, Germans, Kelts, Romans and Greeks. In conclusionmention may be madeof theAlbanian word perend,, perndS " god." If, as has been suggested,

I Mon.Germ., ii, 343. ' If the customaryidentification of "Hercules"with the thuinder-godis correct,Tacitus silduaHerculi sacra (Ann. ii, 12) maybe added. 3 Pliny,N.H., xii,2 (cf.xvi, 4): Phaedrus,Fab. iii, 17. 4 Ovid(Met. 106) calls the oak " Jupiter'stree." I Cf. Servius ad Aen., vi, 772. D 2 36 H. MUNROCHADWICK.-The Oa and theTIunder-God.

thisword is relatedto Lith.perkiivnas, it is likelythat the association may be traced also here. ? 2. The Sanctuaroyof the Th1under-God. The sanctuaryof Zeus at Dodona,near Jannina,was one of the oldest,and ir early times perhapsthe most importantof all the sanctuariesin Greece. In later timesZeus, togetherwith Dione,had a templeat Dodona; but the early authoritiesmention only an oak or grove. The firstreference is in the Iliad (xvi, 233ff):-" 0 lord Zeus of Dodona Pelasgian,who dwellestafar, who rulest over stornmyDodona, around thee dwell the Selloi, thy interpreters,with unwashed feet,sleeping uponi the ground." The statementthat the priestsor Selloi slept on the groundis repeatedby Sophocles(Treac. 1166f),who mentionsalso that the tree stoodin a grove. Oracles weregiven by the rustlingof the leaves- whencethe tree is representedas endowedwith speech-and by the murmnuring of a streaulwhich issued forth from beneath the oak.2 The streamis called by Pliny(N.H. II, 228) "Jupiter'sspring." The pigeonswhich frequented the grove werereg,arded as sacred. Thereare severalpoints of close resemblancebetween the Greeksanctuary at Dodona and the Prussiansanctuary at Romove. In both cases the oak is the dwelling-placeof the thunder-god.3The neighbourhoodof Dodona is famousfor thunderstorms;4 Perkuno manifestshis presence in the thunder. Oracular responsesare givenforth by theoak at Dodona; so also by the sacredoaks of the Prussians. The priests sleep on the groundround the oak at Dodona; the Prussianpriests live in tentsround the oak at Romove. The sanctityattached to the pigeonsin thegrove at Dodona maybe comparedwith the sanctityattached to thebirds and animalswhich frequented the sacred groves of the Prussians. These poilntsof resemblance,however, are not confinedto the Greek and Prussiansanctuaries. If "1the Keltic imageof Zeus is a loftyoak " andif, as Lucan states,the Druidslived in groves.sanctuaries like those at Dodona and Romove mtusthave beenin existenceamong the Kelts. Again the followingpassage from Claudian (De Bello Getico545ff) seems to show that oracularresponses were given by the sacred trees of the ancient Germans:- " Moreover we have encouragementfrom the gods. It is not dreamsnor birds(which guiide us), but clearspeech issuing from the grove." It is also frequentlystated, both in the Northand amongother German tribes, that nioharm was allowed to be done to any livilngbeing in the neighbourhoodof a sanctuary. Again for the spring beneaththe oak parallels may be foundin northernEurope. Thlesacred oak which Ottofound at Stettin(cf. p. 33) had a springbeneath it. Therewas a springin

' Cf. Od., xiv, 327ff.; Esch. Promn.,848ff.; Soph. Trckch., 170f.; 1164ff. 2 Cf. Servius ad Aenz.,iii, 466. For Dodona this is implied by It., xvi, 233ff.(quoted above). In a fragrmentof Hesiod also (No. 156 in Rzach's edition) Zeus is represented as living,in the oak, but this depends on an emelidation(vacTv or valEt lor MSS. vaZoP). 4 Leke, Tramels in YorthernoGreece, 4, 198. H. MUNROCHADWICK.-Tk1e Oak and tAcThunder- God. 37

theneighbourhood of the sacredtree at Upsala.1 It is stated that the natureof thistree was notknown; but like the oak at Romove it was evergreen. It is, perhaps,worth observing that the Upsala sanctuarymay primarilyhave been connectedwith the worship of Thor. At all eventsThor's image occupied the chief positionin the temple. Lastly the world-tree,Yggdrasill's Ash, is likewise representedas evergreen,and standsover the ",springof Fate." The resultsof thisdiscussion may be brieflysummarised as follows: There weresanctuaries of the thunder-godamong the Greeks and Prussians,probably also amongthe Kelts, consistingof oaks standing,within groves. Perhaps the sacredoaks wereoriginally chosen by preference fromi one or otherof theevergreen kinds.2 Roundthese oaks the priestslived undersomewhat primitive conditions of life. Further,there are resemblancesin points of detail betweenthe tree- sanctuariesof theGermans and Slavs on the one hand,and those of the Greeks and Prussianson the other, sufficientlystriking to justifythe suspicionthat similarconditions may once have prevailedin the sanctuariesof thesenations. As regardsthe Germans3perhaps some objectioln may reasonably be takento the assumptionin thelast statement. There is, however,some indirectevidence to stupportit. It has been suggestedabove that in the Northerntemlple the " langhus" is a developmentof the dwelling-house,probably that of the priestor chief'of the community,while the " afhu's" seemsto have takenthe place of the sacredgrove and to bear someclose relationshipto the Vardtradwhich is found standinigbeside the ordinaryhouse. Now thereare tracesthat in earlytimes the relativepositions of thehouse and treewere sometimesdifferent. In the hall of V6lsung'spalace, according, to VdlsungaSaga 2 (Fornald.Sog. I, 119), stoodthe steni of a hugeoak, the branchesand foliageof whichspread out above the roof. A house of this type may obviouslybe a developmentof a tree-dwellingsimilar to the sanctuaryat Rolmiove.I do notknow of any otherexample of a houseof this kind in the North., Perhaps,however, the followingpossibility is worthtaking ilnto account. Can the " high-seatpillars " whichstood within the hall,both of secular and sacredbuildings, and were regardedwith peculiar reverence, owe theirorigin to the formiierpresence of an oak in the sameposition ? It is not stated tllatthey weremake of oak-wood,but fromtheir association with the thunder-godit is likely thatsuch was originallvthe case.0 The figureof Thor whichwas carvedtipon themmay be comparedwith the Slavonicidols discussedabove. Schol.134 to Adamof Bremen. 2 It is worthnoting that the Roman"; civic crowln " was originallyiliade from the leaves of theevergreen oak (ilev); cf.Pliny, N.H., xvi,4. In regardto the Slavs,I have tried to show above (p. 33) that the Slavonictemples in Rgiigenmay come from tree-sanictuaries siinilar to thatat Romove. 4 In theNorth the temporal chief seems to havebeeln also thepriest of the coniultinity. s The descriptionof the thalamos of Odysseus (Od., xxiii, 190ff.) imiay be compared. It is perhapsworth suggesting that the reqin-maqlar, which stood in thepillars, nlay lhave beenpegs used for ignition by frictioni,perhaps for the re-kinldling of the perpetualfire, which, in viewof certaini customs existing in latertimes, may have been exting,uishedonce a year. Or again it niaylhave beeni for the kiindlingof the" iieed-file" which was also perhapscoinnected 38 H. MUNROCHADWICK.-The Oak and theThunder-God.

CHAPTER IV. It has beenshown above: (1) That the cult of thethunder-god can be proved to have existedamong most of the Indogermanic-speakingniations of Europe. He appears to have been the chiefdeity of the Greeks,Romans, and Kelts,in pre- historictimes probably also of the Germans(at all eventsin the North). Further if Procopius'statement (Gothic War, iii, 14), thatin his time(the sixth century) the thunder-godwas the onlydeity worshipped by the Slavs,be comparedwith theprominent position occupied by Perkunoin thereligion of thePrussians, there can be littledoubt that the thunder-godwas originallythe chief,if not the only deityof the Baltic and Slavonic peeples. (2) That amongall thesenations holy treesand grovesare found. The treemost generally venerated seems to have been the oak. (3) That the oak seemsto have been associatedwith the cult of the thunder-godamong the Greeks,Romans, Kelts, Germans, and Prussians; further, that oak-sanctuiariesof the thunder-godshowing, striking features of resemblance are foundamong the Greeks and Prussialls, and thatthere are grounds for suspecting that similarsanctuaries have existedamong the Kelts, Germans,and Slavs. It remainsnow to enquirewhat can have been the originalnature of the association betweenthe thunder-godand the oak.

? I. Mr. Frazer'sTheory. Mr. Frazer holds that the oak was originallynot merelythe symbolor habitationof thegod, but was itselfthe objectof worship. In The GoldenBough, vol. ii, p. 291ff,he writes:-" If then the greatgod of both Greeksand Ronman.s was representedin someof hisoldest shrines under the formof an oak, and if the oak was the principalobject of worshipof Celts, Germans,and Slavs, we may certainlyconclude that this tree was one of the chief,if not the verychief divinity of the Aryansbefore the dispersion." This passage must be readin connection withvol. i, p. 62:- " In thesecases the spiritis viewedas incorporatein the tree; it animatesthe treeand mostsuffer and die withit. But accordingto another, and no doubtlater view, the treeis not the body,but merelythe abode of the tree-spirit,which can quit the injuredtree as men quit a dilapidatedhouse." Also withvol. i, p. 65:-" When a tree comesto be viewedno longeras the body of the tree-spirit,but simplyas its dwellingplace whichit can quit at pleasure,an importantadvance has been made in religiousthought. Animismis passinginto polytheism." Originallytherefore the oak was itselfthe deity; the conceptionof it as the dwelling-placeof the deityis a laterdevelopment. Mr. Frazerseemiis to me to assulmetoo much in his statement(vol. ii, p. 291) that the oak was "not onlythe sacred tree,but the principalobject of worship with the cult of Thor; cf. Adam of Bremen,iv, 26: sipestis uel fames imminetThor ydolo libatur. In later timesthe sparks forthe kindlingof the " need-fire" were sometimesobtained by twisting a woodenipeg rouindin an caken post (cf. Grimm,Deutsche Aythologie,i, 502ff). H. MUNROCHADWICK.-The Oak and theTh7under-God. 39 of bothCelts and Slavs." His authoritiesfor this statementare the passages quotedabove regarding the tree-cultof the Lithuaniansand Prussians,'and Pliny's accountof the cuttingof the mistletoe. But it is clearly stated in many places thatthe Lithuaniansand Prussiansregarded their groves and treesas thedwelling- places of the gods. Again Pliny, though he says that " the Druids esteemnothing more holy than the mistletoeand the tree on whichit grows, providedolnly that this is an oak,"yet adds the followingexplanation for this fact:-" theybelieve that whatever grows on thesetrees is sent fromheaven, and is a signthat the tree has been chosenby the god himself." The god is clearly notinseparable from the treehere. For illustrationsof the originalform of cult, whereinthe treeand the tree-spiritare identical,Mr. Frazerhas to go far beyond Europe.2 ? 2. Objectionsto theabove theory. Mr. Frazer's statementswould lead one to the conclusionthat amongthe Indogermanic-speakingpeoples the developmentof polytheismfrom animism, or at all events the developmentof the thunder-godfrom the oak, took place subsequentlyto the " dispersion." Against this supposition,however, there are severalserious objections: I. Thoughseveral of the Indogermaniclanguages possess wordsfor "god" peculiarto themselves,yet it is practicallycertain that one wordmust have been used with this meaningeven in the parent language. This is shown by the identityof the Indian, Italic, Keltic, Germanic,and Baltic words for "god" (Sanskr.devas, Lat deus,diuus, O.Irish dia, O.Norsetivaqr (pL.),3 Lith.devas). II. No tree-nameis everused for" god,"nor is the thunder-godever denoted by a word whichmay have beenia tree-name.The Baltic-Slavonicdesignatioln of the thunder-godseems indeed to be derivedfrom the Ildogermanicname of theoak (*pek7ktus),but it is a derivativeand not the worditself. Hence it would seem to have beenoriginally an epithet,"having somethingto do withthe oak,"4 perhaps"living in the oak." III. The nameof the thunder-godin Keltic alndGermanic is identicalwith the wordfor thunder. It is probablethat a similarword must once have existed in Baltic and Slavonic. Otherwisethe use of the wordsperk-inas, perun for "thunder,"" thunderbolt"is difficultto explain. On the otherhand the name of the thunder-godin Italic and Greek seem to have originallymeant "sky," "daylight" (cf. Sanskr.dyaus "sky," " day," also personified;Lat. dies" day"). But,further, this word seems to be ultimatelyconinected with Indogerm.*deiuos

I I ouighthere to acknowledge my obligations to Mr. Frazer's book for several of these references.

n I have not the qualifications necessary for enteriniginto a discussion of these examples, but I suspect that some of them might be capable of more than one interpretation. 3 The singular is only used as the proper name of a god: O.Norse, T r; O.H.G., Zio; O.Eiigl., A, 7iw-. 4 Cf. Gk. 4bqycovaiosas an epithet of Zeus (Stephanus Byzantiniusunder Awo&Uwj). 40 H. MUNRtOCHADWICK.-he Oak c(flb(the Th6under-God.

"god." This tends to show that among the Indogermnanic-speakingpeoples the conceptionof " god" in generalwas bound up with that of " sky,"and that this was moreespecially the case withthe conceptionof the thunder-god.It imustbe admlittedthat it is at all events more natural that the thundershould be conceivedof as comingfrom the sky' than as proceedincrfrom an oak or any othertree. These arguimentsseem to me to tell greatly against the probabilityof Mr. Frazer'shypothesis, and in favourof the suppositiolnthat tlle conceptionof the thundler-godwas originallybouind up witlhor developedout of tImeconception of "sky." Thierefore,since it is obvious that the two hypotllesesare miutually exclusive,it renmainsto be seen whetlherthe associatiolnof the thulnder-godwith tlie oak maybe of secondaryorigill. In the followingpages I will endleavouLrto showthat this miiay be tihecase.

? 3. Conclusion. It has been slhowiiabove that in the Greekamijd Prussian sanctuaries of the tllunder-godthe priestslived beneaththe sacred t;ree,and that there is sonie reason for supposinig that the same custoiii may once have prevailed among the Kelts,Germans, and Slavs. One might,perhaps, say " chiefs" for" priests,"for in the earliesttimes it is probablethat timetwo officeswere united. Such appears to have always been the case in the North,2while among the, Prussiansit is noteworthythat Bruteno,the legendaryfirst high-priest, had formerlybeen king, The questionmust now be asked: Is it necessaryto supposethat the priests lived beneaththe oak because it was sacred? Is it not possible that the oak acquiredits sanctityfrom the factthat the priestslived beneathit ? Accordingto RobertsonSmith (Religionof the Sernites,'p. 197) the oldest sanctuariesamong the Phceniciansappear to have been naturalor artificialcaves. He explains thisfact by suggestingthat in this districtcaves were the earliest formof humanhabitation. Now what is likely to have been the earliestformii of humlanlhabitation in the more northerniparts of the continentof Europe? Over a considerablepart of the area, which in the earliesthistorical times was occupied by Qermansandl Slavs, caves would not oftenbe available. On the otherhand it is klnowlnthat great portionsof these coniltrieswere at one time alniostcomi)pletely covered with forest. Is it notprobable that in primitivetimes the inhabitantsof these regionsmade theirhome under the shelterof the larger foresttrees ? Thoughthis maybe granted,it will,perhaps, be said that such a stateof societymust lie too farback in the past for any reminiscencesthereof to have survivedin historicaltimes. Yet accordingto ErasmusStella' the Prussians

I Cf. the Homeric phrase ZEvu ai'&pt vaiecv(II. ii, 412), and Adam of Bremeln,iv, 26: Tlhorpraesidet in aere. 2 I hopeto deal withthis quiestioln, so faras the Germanicnations are concerned,oil some futureoccasion. 3 Grynaeus, op. cit.,p. 574. H. MuN1RoCHADWICK.-TICe Oak and thleThunder-God. 41 had a traditionthat at one time they didl not build houses but protected themselvesand their childrenfrom rain and cold by eaves and by the bark of trees. Iordanes (e. 5) says that the Slavs inhabitedforests and marshes ilnsteadof cities. The dwellinigsof the Germansin the firstcentury and of the Slavs in the sixthcentury, as describedby Tacitus and Procopiusrespectively, seem to have been of a somewhatprimitive description. But accordingto Tacitus the Finns in the firstcentury were still withouthouses. Theyslept on the ground,and theirhomes consisted simply of a networkof boughs'with which theyprotected their ehildren and old peoplk. Now thecustoms of a moreprimitive stateof societyare oftenpreserved ill sacraluse when the communityin general has reacheda highergrade of eivilisation. The lifeof the priestsat Dodona is practicallyidentical with that ascribedby Tacitus to the Finlns. It may be conjecturedtherefore that this manner of lifewas once practisedby the alncestors of the Greekrace in general. The sanctityof the oak has yet to be explainied. There is reason for believingthat the oak wvasonce the commonest,as well as perhaps the largest tree.in the forestsof northernEuirope. As such it wouldnaturally be chosenfor the habitationof the primitivecommunity and consequentlyof all their belongings,their animals, their guardian spirits and theirtribal god.2 Evergreen trees,such as the oak at Romoveor the tree at Upsala,would obviously have the preference. Subsequently,when the art of building had become known,the tree-homiiewas desertedfor the purposesof ordinarydwellinig, but the sanctityof old associationsclung to it, and it continuedto be regardtedas tIle lhomeof the tribalgod, as Tacitus (Germn.39) says of the groveof the Seminones:tanquacum iizdeinitia gentis, ibi regncatoromniumn deus. Between the primitivecommnunity beneath the oak and the Prussian sanctuaryat Romiovean intermediatestage inaybe traced. Herodotus(iv, 23) relates how far to the northof the Saythians,at the foot of loftymountains, there dwelt a race of holy men called Argippaioi,each under his own tree. This tree was uncoveredin summerbut in winterwas surroundedwith ain impenetrablecovering of whitefelt. It is interestingto observethat, like the 1)ruicls these patriarchswere not onlyregarded as sacred,but were also appealed to as judges in all cases of dispute. Their homes were inviolable places of asylunmand, like the priestsof the ancientEnglish, they possessed no weaponsof war. Does the curtainedtree-dwelling of these patriarchsrepresent an earlier formof the type seen at Romove? If so it is a natural inferencethat the curtainwas originallyno more than a protectionfor the primitivecommunity againstcold and wet.

1 These dwellings would presuinablybe somewlhatsimilar to the rough shelters still in use amonlgtlle dwarf yellow races of Equatorial and Southern Africa. 2 To this is probably to be added the fact that in primitive times the acorn seems to have beeni a, common article of food. The traditioinwas preserved by both Greeks anld Romans, cf. Frazer, op. cit., ii, 1). 292, footnlote,anid the referenicesthere quioted. 42 H. MUNROCHADWIcK.-The Oak and theThunder-God.

To sum up brieflythe resultsof this discussion,I am of opinionthat the thunder-godwas supposedto inhabitthe oak becausethis had formerlybeen the dwelling-placeof his worshippers.Originally, no doubt,he was conceivedof as dwellingin the sky; but fromthe very close connectionwhich exists in all primitivecommunities between the god and his people,it becameinevitable that he shouldbe regardedas presentin the home of the co'mmunity.When the communitytook to building and desertedthe tree-home,the sanctityof old associationsclulng to the latter,and the god was still supposedto dwell there. This is the stage of societyrepresented by the Germansof Tacitus'day and by the Pruissiansup to theirconversion.' The protectionof the god over the new homewas obtained,in theNorth at all events,by the importationinto it of a pillar (probablycut froma holytree) with the imageof the god carvedupon it. The thirdand last stagewas reachedby the accomnmodationof the god in a temple builtlike human habitations, but withcertain peculiarities which may be due to reminiscencesof thegrove-sanctuary. This is the stagefound in theNorth in the last days of heathendom.The change,however, was not complete,for, in certain cases at all events,the sacred treeor grovecontinued to existby the side of the moremodern temple. It maybe objectedto thisexplanation that it is in no wayspecially applicable to thecult of thethunder-god. Against this I wouldanswer that the thunder-god was thegod of the primitiveEuropean community.He is the only god whose ciultis commonto the Indogermanic-speakingpeoples of Europe. It is probable that these peoples,like the ancientSemites (cf. Robertson Smith, op. cit.,p. 39), thoughthey recognised the existenceof manysupernatural beings, were yet not polytheistsin the sense that theyworshipped nmore than one god. Accordingto Procopiusthe Slavs of the sixthcentury had onlyone god,namely, the thunder- god, thoughthey also worshippedSatpovta. The same was probablyalso the case in the Northin earlytimes. The cult of Fro seemsoriginally to have been peculiar to Sweden and, togetherwith that of Ni6rdr,is capable of a special explanation. The worshipof Othin was probablyintroduced at a comparatively late period. No othercult seems ever to have attainedmuch importance in the North. Accordingto RobertsonSmith (I.e.) pantheons,such as the Greek,do not belongto the primitivestages of society. Each communityhas its god (and perhaps a goddess),its guardian spirits and tribalheroes. In the primitive Europeancommunity the god seemsalways to have been the thunder-god.

DIscuSSION. Mr. W. GOWLAND,after complimenting the authoron his extremelyinterest- ing paper,called attentionto severalanalogies between the sacrednessof treesin earlyand latertimes in Europeand in the Far East-in Japan and Korea. I It is noteworthythat the Prussians took their dead to the sacredwoods, where, according to Michov.each familyhad a hearthspecially reserved to it forcremation (Grynaeus, op. cit., p. 520). H. MUNRO CHADWICK.-The Oak and the Thutnder-God. 43

In Japan in mostvillages there is a sacred tree,the actual tree,however, not being reverencedper se, but as the abode of some god or spirit. It was not reg,ardedas the homiieof the spirit,for this is always in the sky,but miierelyas his terrestrialdwelling-place, vliere he could receive the offeringsand reverence

SACRED PINE TREE IN A JAPANESE VILLAGE.

which the inlhabitanitsof thie village might wish to offer. The spiritwas often held to be the peculiar guardian of the prosperityof the village, even when, as oftenhappened, his nature was malevolent. 44 H. MUNROCIIADWICK.-The Oak-ani(d the Thunder- God.

The treewas alwayselicireled by a strawrope with pendantsof straw or of strawand paper suspendedfrom it resemblinga fringe,as shownoln p. 43. This is a perfectparallel of the curtain mentionedby the author. The offerings consistedof foodaud wine,frequently also of rounidwater-worn stones, the latter beingpiled up aroundthe base of the trulnk. Thesesacred trees belong essentially to Shint6spirits and the offeringsmade to themare evidentlysurvivals of an ancienfpagan cult. They are sometimes foundin theimmediate vicinity of a Shintotemple, and occasionallysmnall ShintP3 shrinesare fixedon their trunks. The tree in South and Central Japan is generally the pine (Pinits Thmnbergii') but. sometimes the canmphortree (Cinna- momunqmCamphora), not because thereis any special virtuein either tree,but becausethey are the largesttrees in the district. In Korea he had foundthe samepractice. The sacredtree was the largestin the neiohbourhood.Its trunkwas encircledby a rope with pendantsof strawor ofstrips of rags. Heaps ofwater-worn stones offered by the villagersor bypassing travellerswere piled up at its base. Grovesof trees were not regardedas sacredin Japan,and templeswere not builtto resemblegroves, but everyimportant Shint6 temple was erectedin a grove, or in froiitof a clumpof treeswhich were often of gigantic size. Mr. G. M. ATKINSON said that he would be glad to knowif therewas any connectionor continuitybetween the GospelOaks and the ThunderGod, and why the oak leaves are worn on the 29th of May, for the popular explanationis unsatisfactory.The battle of Worcesterwas foughtin September. He thought theirstructural ornamentation went far to provethe timberorigin of the Temples. The ritual of the rag-bushand its connectioniwith NorthernMythology has beelnexplained by Mr. M. J. Walhouse; but it is Yggdrasill'sash-tree that comesin forthe greatest share of its attention-