andreas Nordberg
a society can only function if there is a common perception of and a more or less agreed-upon system for the calculation of time. for the individual and family, time is governed by mundane and periodic tasks. for settlements and larger regions, time regulates collective social arrangements, political, judicial and economic harmony, and, not least, both private and public religious life. The anthropologist Edmund leach has expressed it thus:
among the various functions which the holding of festivals may fulfil, one very important function is the ordering of time. The interval between two successive festivals of the same type is a ‘period’, usually a named period, e.g. ‘week’, ‘year’.
Without the festivals, such periods would not exist, and all order would go out of social life. We talk about measuring time, as if time was a concrete thing waiting to be measured; but in fact we create time by creating intervals in social life. Until we have done this, there is no time to be measured. (leach 1961: 134–35) another anthropologist, Victor Turner, described time in religious societies as a movement between liminal and non-liminal periods (Turner 1969), the liminal constituting the central phases of the ritual performances. as we have already dealt with the various phases in (è25), this ‘ritual time’ in the narrow sense will not be treated in this chapter, which will be concerned mainly with the calendar that can also be seen as a sort of ‘ritual calendar’.
The oldest extant source materials concerning possible traces of a pre-Christian division of time in the Nordic region originate in twelfth-century iceland, but some scant evidence is also attested in other parts of scandinavia.
Andreas Nordberg, associate Professor of the History of Religions, stockholm University The Pre-Christian Religions of the North: History and Structures, Volume ii, ed. by Jens Peter schjødt, John lindow, and anders andrén, PCRN-HS 2 (Turn hout: Brepols, 2020) pp. 725–738
BREPols
PUBlisHERs
10.1484/M.PCRN-EB.5.116955
726
Andreas Nordberg
We cannot assume that the reckoning of time was exactly the same across this whole area, but we can nonetheless surmise that there were basic similarities in the calendrical divisions, since these periods and points of demarcation to a large extent depended on elementary astronomical and climatological divisions. logically, the day was divided into periods based on observations of, for example, the sun’s passing across the heavens and its rising and setting on the horizon. longer periods were divided into weeks, months, seasons, and years.
Months followed the four phases of the moon (new moon, waxing gibbous, full moon, and waning gibbous), while the year could be divided both climatologically into seasons or astronomically in accordance with pivotal points such as the solstices and equinoxes.
although there is no doubt that the measurement of time would, of course, have varied in its details from one region to another, there is data to indicate that efforts were made in pre-Christian society to establish some degree of uniformity on how time was measured. according to the icelandic code Grágás, for example, during the althing, the law-speaker should proclaim the correct calendar, after which the icelandic goðar (plural) should communicate this to the local legal assemblies (Hastrup 1985: 27–28) further, it was stipulated in several swedish and Norwegian provincial laws that it was the clergy’s duty to explain the calendar to the masses (granlund and granlund 1973: 21–22 with refs.) similarly, it may have been the duty of the law-men and religious leaders to maintain and communicate the calendar in pre-Christian times. ** **
Solar Year and Lunar Months
The gregorian calendar, which is used in the West today, is a modification of the Julian calendar initiated by Julius Caesar in the year 45 bce as the official calendar of the Roman Empire. When Christianity was adopted as the official religion of Rome, the Julian calendar became the official calendar of the bur-geoning Christian Church. The spread of Christianity throughout Europe thus brought with it the Roman way of measuring time. in the Nordic areas, the Julian calendar seems gradually to have replaced two different ways of measuring time: on the one hand, a form of weekly calendar; and on the other, what is usually referred to as a ‘lunisolar’ calendar. This latter calendar was based on the solar year, while the months were calculated from the phases of the moon, probably from new moon to new moon. at least some of the annual festivals, markets, and legal assemblies seem to have been regulated on the basis of this lunisolar year. in many ways, this is quite understandable. The lunisolar calendar was based on the astronomical cycles of the sun and moon, and these are
28 – Ritual Time and Time Reckoning
727
constants no matter where one is located. Thus, placing a market, a legal assembly, or a religious festival in accordance with a certain phase of the moon in a specific lunar month would make coming dates for that gathering very easy to remember and to calculate.
since a synodic lunar month lasts for approximately 29½ days, a lunar year has only 354 days. This causes a dislocation of eleven days between the lunar and the solar years which seems to have been compensated for by insert-ing a thirteenth month approximately every third year (see Nordberg 2006a: 51–75 with refs.). There is much to indicate that the point of departure for this system involving intercalary months was the winter solstice. in Chapter 15 of Bede’s De temporum ratione from 726 ce, there is an account of a lunisolar year which was apparently employed by the heathen angles (probably also the Jutes and saxons). The first month of the year, corresponding approximately to January, was called Giuli ‘yule-month’. This month was followed by Solmonath, Hredmonath, Eosturmonath, and Thrimilchi. The months corresponding to June and July were both called Litha. Then followed Weodmonath, Halegmonath, Winterfilleth, and Blodmonath. as January, also December was called Giuli. The year was further divided into four quarters of three months each. leap years consisting of thirteen months were called Thrilithi, because a third Litha- month was inserted into the summer of these years. The year started at the winter solstice, which, according to Bede, was called Modranecht ‘the night of the Mothers’, after certain religious festivals which took place at that time, and in addition, the two lunar months Giuli were named after the winter solstice, since one preceded and the other succeeded that day. further, Bede explains that Solmonath could also be called ‘the month of cakes’, since cakes were offered to the gods during that month. Hrethonath and Eosturmonath were named after their respective goddesses, Hretha and Eostre. in Thrimilchi, cattle were milked three times a day. Litha meant ‘gentle’ or ‘navigable’, referring to the calm breezes of the sea during that part of the year. Weodmonath meant ‘month of tares’ and Halegmonath ‘month of sacred rites’. Winterfilleth alluded to the first full moon of the winter. according to Bede, Blodmonath (literally ‘blood-month’), meant ‘month of sacrifices’ because the cattle slaughtered during this month were consecrated to the gods.
similar accounts of these old names of months, albeit less detailed, are to be found in other old English sources as well. an incomplete calendar of Normanno-saxon character is found in Cotton Vitellius E. XViii. The manuscript, which is believed to have been composed in the year 1031, contains the following series of months: solmonað (february), Hlyda (Mars), aprelis monað, maius, junius or ærra liða, julius monað, weodmonað (august), halig-
728
Andreas Nordberg
monað (september), winterfylleð (october), blotmonað (November), and ærra jula (December). Many of these names are also accounted for in a series of antique names of months, compiled 1705 by g. Hickes in Antiquæ litteraturæ septentrionalis libri duo i. for example, the double months associated with the summer solstice are here called ærra liða ‘the earlier liða’ (= June) and æftera liða ‘the later liða’ (= July). similarly, December and January are recognized as ærre geola and æftera geola ‘the earlier/later yule month’ (Nilsson 1920: 293–
94).The earlier yule month of these series even has an early gothic parallel: fruma Jiuleis ‘the month before yule month’, confirmed in the fragment Codex Ambrosianus from about 350 ce. The name of the month implies, as in the old English accounts, a lost second month * Jiuleis (feist 1923).
The lunisolar calendar described by Bede was obviously closely related to the ritual year. Due to the lack of relevant sources, it cannot be established whether a similar relationship between the naming of months and religious conceptions and rituals once existed also in pre-Christian scandinavia, but many indications still suggest that the lunisolar calendar was of great importance for the ritual year.
With some important exceptions (see below), the names of the pre-Christian lunar months from mainland scandinavia are no longer known. The oldest known list of names comes from early medieval iceland, and there are two main sources for it: snorri’s Edda and the text Bókarbót, dated to c. 1220. in these writings, the months are part of an exclusively icelandic medieval calendar. The summer half of the year contains the following names, beginning with a month stretching from mid-april to mid-May: Gauksmanuðr ‘cuckoo month’ or Sáðtið ‘seed-time’ or Harpa (meaning unknown). Mid-May to mid-June: Eggtíð ‘the time when birds lay eggs’ or Stekktíð ‘the time when lambs were separated from their mothers and put into enclosure ( stekkr)’ or Skerpa (meaning unknown). Mid-June to mid-July: Sólmánuðr ‘month of the sun’ or Selmánuðr ‘shieling month’. Mid-July to mid-august: Miðsumar ‘midsummer’
or Heyannir ‘hay-time’. Mid-august to mid-september: Tvimánuðr ‘the second month before the beginning of winter’ or Kornskurðarmánuðr ‘barley-cutting month’ or Heyannir. Mid-august to mid-september: Kornskurðarmánuðr ‘barley-cutting month’. Mid-september to mid-october: Haustmanuðr ‘the month of harvest’. The winter half of the year contained the following months, starting with mid-october to mid-November: Gormánuðr ‘slaughtering month’.
Mid-November to mid-December: Frermánuðr ‘frost-month’ or Ýlir (cognate with jól ‘yule’). Mid-December to mid-January: Jólmánuðr ‘yule-month’ or Mǫrsugr ‘fat-sucker’ or Hrútmánuðr ‘ram month’ (with reference to the pairing of the sheep). Mid-January to mid-february: Þorri (meaning unknown). Mid-
28 – Ritual Time and Time Reckoning
729
february to mid-Mars: Gói (meaning unknown). Mid-March to mid-april: Einmánuðr ‘the last month before the beginning of summer’ (Nilsson 1920: 297–98; Beckman 1934: 32–34).
according to Martin P:n Nilsson and Nathanael Beckman, it is doubtful whether all of these names were actually commonly used, except for Jólmánuðr, Þorri, and Gói which correspond to names of lunar months in other parts of scandinavia. in Norway, finn Magnusson recorded the names Jolemoane, Thorre, the latter used variably for January and february, and Gjö, variably for february and March. The variation in the accordance of these months with those of the Julian/gregorian calendar is explained by the fact that the Norwegian names were names of lunar months (Nilsson 1920: 298). in medieval and later swedish sources, there are several fragments of a lunisolar calendar.
among the lunar months are testified, for example, Julmånad/ Jultungel, Tor/
Torre, and Göje/ Göja.
in early medieval iceland, the month Jólmánuðr was said to last from mid-December to mid-January in the Julian calendar. it was preceded by the month Ýlir, the name being an ia-derivation of old Norse jól ‘yule’, which corresponded to the period mid-November to mid-December. The winter solstice occurred at the junction of these two yule-months (Beckman 1914: 169). it is thus usually accepted that all of these month-names originally designated pre-Christian lunar months which, like the anglian and old English yule months, were in some way connected to the winter solstice. Possibly, the basic rule for calculating this lunisolar calendar was that the crescent of the second yule lunar month was not to be seen in the sky before the winter solstice, and that a thirteenth leap month was to be inserted the present year, if this was about to happen the following year (Nordberg 2006a: 65–66).
The Names of Days and Calculation of Weeks
over and above calculations based on the lunisolar year, it appears that time was also measured on the basis of week years. according to a report which deals primarily with the establishment of the alþingi in iceland in 930, the icelandic year at that time was considered to consist of fifty-two seven-day weeks (Hastrup 1985: 25–26). it is commonly thought possible that the icelandic settlers brought this weekly measurement from scandinavia. its origin is, however, still quite uncertain.
Etymologically, the word for ‘week’ (old Norse vika; old swedish veka, vicka; old High german wecha; anglo-saxon wice) can be traced back to
*wikōn ‘change’ and may originally have referred to the four roughly seven-day-long phases of the moon (Hellquist 1957: 1324). although it is possible that
730
Andreas Nordberg
the week in this sense could be traced back to an ancient indigenous tradition, the names of the weekdays in the germanic languages are, in fact, reinterpretations of the Roman weekdays. The day of the sun (old Norse Sunnudagr, old English Sunnendæg, old High german Sunnûntag, etc.) have their equivalent in the dies Solis of the Roman week, while the day of the moon (old Norse Mánadagr, old English Monandæg, old High german Mânetag) corresponds to the latin dies Lunae. The god Týr’s day (old Norse Týsdagr, old English
Tiwesdæg, Middle High german Zîstag) are the equivalent of the Roman god Mars’s day, dies Martis, while Óðinn’s day (old Norse Óðinsdagr, old English
Wodnesdæg, Middle High german Gudensdach) is modelled on the latin dies Mercurii, Mercury’s day. Þórr’s day (old Norse Þórsdagr, old English Þunresdæg, old High german Donrestag) equate to Jupiter’s day dies Iovis, and frigg’s day (old Norse Frjádagr, old English Frigedæg, old High german Frijatag) can be traced back to Venus’s day dies Veneris. only the latin name for saturday, Saturni dies, lacked an old germanic equivalent (green 1998: 244, 248–53; see è13).
There is much to suggest that these day names were absorbed through interpretatio Germanica by the germanic peoples of the Continent and the scandinavian peninsula through contact with the Roman Empire in the first centuries ce and were then incorporated into a calendrical system, which may have been a precursor of the Nordic week year. if such is the case, one might be tempted to see the origin of the week calendar as part of a larger process of change among the germanic peoples wherein social organization, trade, systems of justice, the art of war and runic characters, and so forth were developed or reformed partly based on Roman models (see è13).
There is, however, no exact information on how this pre-Christian week calendar was ordered. When measurement by weeks emerges in greater detail in the icelandic sources in the twelfth and thirteenth centuries, it is partly integrated into the ecclesiastical measurement of time. in scandinavia and in the old swedish settlements in Estonia and Karelia from the seventeenth century onwards, it is found only as calendrical fragments. incomplete remains of information are even to be found in silesian sources from the sixteenth century. What is best preserved is what appears to be a closely related week calendar, which was still being used in sámi society as late as the nineteenth century (lithberg 1944; granlund 1955; granlund and granlund 1973; Vilkuna 1974: 279–80). Judging from these remnants, the week measurement was tied to important periods in the working year. similarly, the measurement by weeks can indirectly have been of great importance even to the annual cycle of festivals and can have constituted the calendrical starting point for the division of the year into four quarters.
28 – Ritual Time and Time Reckoning
731
Division into Quarters and Annual Festivals
Calendrical festivals definitely existed in several forms. some were of only regional or even local importance. it is quite possible, for example, that the so-called álfablót and the blót in Vǫlsa þáttr, celebrated in late autumn at local farms or villages,1 are two known examples of such local or regional festivals.
according to the scant sources concerning these two festivals, they seem to have been unfamiliar to strangers. other yearly festivals were undoubtedly celebrated across much wider areas. among the most important were certainly those celebrated at the beginning of the calendrical quarters of the year.
There seem to have been at least two principal ways of dividing the year into quarters. The week year was divided into four quarters of thirteen weeks each. in the case of a peculiar medieval icelandic week measurement, quarters were calculated with the help of the regulations of the ecclesiastical computus.
later scandinavian sources maintain instead that the quarters often coincide with the quarterly divisions of the Julian calendar. in this case, the week year started with Christmas Day, 25 December, which was the formal winter solstice in the Julian calendar. The second quarter started on lady Day, 25 March, which was the spring equinox according to the Julian calendar, while the third quarter started at the summer solstice on the feast of John the Baptist, 24
June. formally, the fourth quarter started on 24 september, but occasionally it was counted from Michaelmas, the 29th of the same month. Many have suggested that this division may be a medieval harmonization of an older quarterly division, which originally followed the astronomical solstices and equinoxes (granlund 1955: 30–32).
There are, moreover, reminiscences of yet another pre-Christian quarterly system in iceland, Norway, sweden, and in the swedish settlements of finland and Estonia, which seems to have been of significance to both the judicial and religious system. These dates are first mentioned in icelandic calendrical texts from the twelfth and thirteenth centuries, but are also found in several scandinavian provincial laws and mediaeval Church calendars, on calendar rods and rune-staffs, and in later popular contexts from the seventeenth century onwards. according to this system of division, the year began in the autumn on the ‘first day of winter’ or ‘the winter nights’ (old Norse vetrnǽtr, old swedish wintirnætir), which in the extant sources is variably given as one of the three days, 13–15 october in the Julian calendar (20–22 November in the 1 in Vǫlsa þáttr the ritual celebrations also takes place on a daily basis during the autumn (see also è31).
732
Andreas Nordberg
gregorian calendar used today). it has rightly been pointed out that this three-day period was originally conflated. The second quarter began at ‘mid-winter’ or ‘mid-winter night’ (old Norse miðvetr, miðvetrarsnótt), 12–14 January in the Julian calendar (gregorian 19–21 January). The beginning of the third quarter was on ‘the
first day of summer’ or ‘the summer nights’ (old Norse sumarmál), 13–15 april (gregorian 20–22 april), which was succeeded by
‘midsummer’ (old Norse miðsumar) 13–15 July (gregorian 20–22
July) (cf. lithberg 1921; Nordberg 2006a: 34–42 with refs.). it has been postulated that these displacements of the quarters in correla-tion to the astronomical solstices and equinoxes were relative to the climatological seasons of the central areas of scandinavia (Vilkuna 1961: 80–83). still, they also seem to have been fixed exactly four weeks after the astronomical solstices and equinoxes, which would imply that the fixed dates of the quarters originally belonged to the old week calendar (Nordberg 2006a: 42–47).
at all events, the displaced quarters were, according to several sources, also tied to the yearly cycle of festivals. Most pre-Christian festivals were probably of regional or even local importance (compare, for example, the alfablót and the blót in Vǫlsa þáttr mentioned above). But some seem to have been more widely dispersed.
in Ynglingasaga ch. 8, snorri sturluson relates that in pre-Christian times sacrifices were to be held at the approach of winter for the year’s crops (‘blóta í móti vetri til árs’), in mid-winter for the harvest (‘miðjum vetri blóta til gróðrar’) and on a third occasion close to summer ( at sumri). He speaks of similar sacrifices in several other texts: for example, in Óláfs saga helga ch. 107–09 where he mentions a sacrifice for good harvest at the winter nights ( at vetrnóttum), a sacrifice for peace and fertility ( til friðar) at midwinter and a sacrifice at the beginning of summer ( at sumri). in several sources, the festival at the winter nights is associated with the female fertility goddesses called dísir, as in Víga-Glúms saga ch. 6 (cf. gunnell 2000), although occasionally also freyr is mentioned as the recipient of offerings at the winter nights ( Gísla saga Súrssonar ch. 15). This might indicate that the cult at the beginning figure 28.1. Runic calendar staff from from Nyköping in södermanland, dated to the thirteenth century (sHM 29486:120336).
Photo: Bengt a. lundberg, statens Historiska Museum, stockholm.
28 – Ritual Time and Time Reckoning
733
of winter included a hieros gamos between the dísir and freyr, as suggested by folke ström (1954). The mid-winter festival appears to be identical with the pre-Christian yule. according to, for example, snorri’s Hákonar saga góða ch.
13 the yule festival fell on ‘ hǫkunótt, it was miðsvetrarnótt and yule was held for three nights’ (see also è31).
There is much to support the idea that these festivities did not coincide exactly with the dates of the year’s quarterly divisions, but rather with a new or full moon connected in one way or another to these dates although moving in the solar year. in his Germania ch. 11, the Roman historian Tacitus mentions that the germanic peoples held their gatherings at a new or full moon, and he intimates in his Annals 1.50 a larger germanic festival at the astronomical new moon, saying that on the night of this festival only the stars lit the sky. similar accounts are found even in later Nordic sources. for example, in Egils saga Skallagrímssonar ch. 44 we are told that a sacrifice to the dísir was held in the autumn when it was niðamyrkr outside. The term describes the darkness when the moon is completely invisible before the first crescent of the new moon is seen in the sky. furthermore, in medieval iceland it was customary to welcome
the new moon in the months of Þorri and Gói, inviting it to shine as a blessing on all things, mankind and animals. similar traditions are also attested in other parts of scandinavia (Celander 1950).
The pre-Christian yule probably involved a series of festivities, but there is also much to support the view that its high point was associated calendrically with a specific phase of the moon, that is, the full moon of the second yule lunar month. if one accepts that the second yule month began with the first new moon after the winter solstice, and that the yule festival was celebrated at the time of this month’s full moon, the varying dates of this full moon would fall within a period coinciding approximately with January. according to snorri, pre-Christian yule was celebrated at Miðsvetrarnótt, which falls right in the middle of that period. further, according to a statement of Thietmar of Merseburg, the most important festival among the heathen Danes was held in January (Nordberg 2006a: 102–07). Both of these festivals accordingly coincide with the second yule month’s full moon, considering that this month began with the first new moon after the winter solstice.
similarly, the timing of the great pre-Christian sacrifice in Uppsala with its market and legal assembly was probably fixed according to the full moon within the lunar month Göja or Göje (old Norse Gói). in several medieval swedish sources, this gathering is named Distingen ‘the dis-councils’,2 and it is said to 2 Despite the similarities in names, the festival Distingen should not be identified with
734
Andreas Nordberg
begin with the first full moon after the first new moon after Epiphany. in medieval and later swedish sources, this moon was named distungel or distingstungel (swedish tungel, meaning ‘moon’ and ‘lunar month’). The use of Epiphany as a starting point for this calendrical rule is probably a medieval alteration of an older pre-Christian regulation. according to snorri’s Óláfs saga helga ch.
77, the gathering was in pre-Christian times held in the month of Gói but was moved to an earlier time of the year when the svear converted to Christianity
— most probably to prevent it from coinciding with the Christian lent prior to Easter. although snorri might have identified Gói with a month in the peculiar and exclusively icelandic calendar of his own time, the name most likely corresponds to the swedish lunar month Göje or Göja. This would have been the third lunar month after the winter solstice, whose full moon occurred in a period roughly corresponding to the month of March. according to adam of Bremen (skolion 141), the main sacrifice of the svear was held in Uppsala ‘circa aequinoctium vernale’ (around the spring equinox). in adam’s day, the spring equinox fell on 15 March, which suggests that adam and snorri in fact write about the same pre-Christian gathering (Nordberg 2006a: 107–15).
a similar timing may possibly even be associated with some other pre-Christian or medieval swedish councils and market assemblies. on the island of frösön (freyr’s island) in the province of Jämtland, the assembly Jamtamot (attested 1170) was held at the spring equinox in the Middle ages (Holm 2000: 66) and during the same period the so-called Samtingen in strängnäs commenced on the first sunday after lent. The name of a lunar month recorded in the seventeenth century, Samtingz tungel, could indicate that also the Samtingen originally took place at a full moon at the spring equinox (lithberg 1944: 144–45, 147).
Longer Calendrical Feast Cycles
over and above such astronomical cycles as the day, the lunar month, and the year, it would appear that the measuring of time and the calendrical cycle of feasts were also associated with a longer astronomical cycle of eight solar years which, with an aberration of one and a half days, constituted exactly ninety-nine lunar months. in practical terms, this means that a lunar month which the West Nordic dísablót. The latter, which took place at the winter nights in autumn, seems to have been a household festival. The Middle swedish Distingen, which was held in spring and encompassed legal assemblies and markets in addition to the religious rituals, was as far as we know the most comprehensive and large-scale public gathering in Viking age svetjud.
28 – Ritual Time and Time Reckoning
735
started at the winter solstice at the beginning of an eight-year cycle would not again be synchronized with the solstice for the following eight years. on the ninth year an approximate synchronization would occur — which was then the first year of the coming eight-year cycle. This cycle has been used in several different ceremonial and calendrical connections, for example, in ancient greece, where it was called oktaëteris.
in scandinavia, this cycle seems to have constituted the basis for the lunisolar calendar and in connection to this to have been associated with at least two major festivals. Thietmar of Merseburg relates of an important sacrifice among the heathen Danes, which took place in lejre ‘post Viiii annos’ (every ninth year) ( Chronicon 1.17) in the month of January. Thietmar’s statement that the sacrifices took place in January agrees well with the interval for the varying dates of the second yule month’s full moon. The expression ‘every ninth year’
is an example of the so-called inclusive way of reckoning, which was common to both latin and demotic languages at this time. in practice, it means that the last year of an interval is included as the starting year of the next interval and is thus counted twice (1–9, 9–17, 17–25, etc.), with the result that, with a modern exclusive way of reckoning, the nine-year cycle only lasts for eight years. it is highly probable that this eight-year sacrificial cycle was identical to the moon’s astronomical cycle of eight solar years and that this also constituted the foundation of the lunisolar year.
if so, the sacrifices carried out at this kind of calendrical festival might even have been imbued with cosmologic-astronomic symbolism. The part of the text in which Thietmar relates of the sacrifices of animals and humans in lejre is from a source-critical point of view problematic in many ways. still, in connection with his information on the amount of sacrificial victims, he mentions the number of 90+9. Coincidentally, 99 is also the number of lunar months encompassed in the moon’s astronomical eight-year cycle (Reuter 1934: 483–
86; Nordberg 2006a: 80–85, 106–07).
Possibly, a similar kind of cosmological-astronomical sacrificial symbolism was implied in connection with the major festival in Uppsala as well. in his Gesta Hammaburgensis (4.27) from the mid-1070s, adam of Bremen relates that this sacrifice was held ‘post novem annos’ (every ninth year, that is, in an eight-year cycle) and that it proceeded for nine days. according to snorri ( Óláfs saga helga ch. 77), it was also customary to hold legal assemblies ( þing) and markets ( markaðr) at this gathering. snorri states explicitly that the market ( kaup-stefna) lasted for a week. according to the ‘Þingmalæ balk’ 14 of the medieval provincial law Upplands-lagen, Distingen’s market days were enclosed by two
‘market councils’ ( kiöpþingæ), the first announcing and the second revoking the
736
Andreas Nordberg
disæþinx friþer ‘Distingen’s peace’. accordingly, the gathering may have lasted for 1+7+1 = 9 days, that is, exactly the period stated by adam ( figure è26).
But since legal councils and trading probably did not occur simultaneously and at the exact same locations as the major sacrifices, the sacrifices were most probably performed in the evenings and during the nights, after the opening legal assembly and after each of the seven market days. on the second assembly, however, the disæþinx friþer was revoked and the meeting was closed. This would suggest a nine day long gathering, which encompassed only eight evenings and nights involving sacrifices. according to adam, nine humans and male animals of different species (nine of each species) were sacrificed, the number of sacrificial victims totalling seventy-two. This suggests that nine victims were sacrificed during each of the eight evenings or nights (9 × 8 = 72).
Possibly, each of these eight nights of sacrifice represented a year in the moon’s (next?) eight-year cycle (Nordberg 2006a: 86–97).
Calendrical Festivals til árs ok friðr
in most religions, time has a mythological-cosmological dimension. This cosmological dimension could also be expressed in ritual. Time and the measuring of time were consolidated through the calendrical festivals. such festivals were held at the beginning of the year, in relation to the beginning of the year’s quarters and in connection with longer cycles important to the reckoning of time. on many occasions it is said of these festivals that sacrifices were held til árs ok friðar (for year and peace). The old Norse word friðr denotes peace and harmony in society, but the word also had sexual connotations and was thus linked to the notion of fertility. The old Norse term ár means ‘year’s crop’, ‘yield of crops and livestock’, and so forth, but its basic meaning is ‘year’
(ström 1976; Hultgård 2003a). The sacrificial symbolism of the cult til árs ok friðar can therefore be seen as reaching further than merely the hope of peace and a good harvest: it also touched upon the wish for cosmic regeneration and the assurance of a new year (cf. è25).3 Therefore, it is no surprise that these practices coincided with the calendrical rites. Time was a crucial aspect of creation.
Noticeably, the most important fertility festivals took place during the winter half of the year; in the beginning of autumn, in the middle of winter, and in in the beginning of spring. This surely depends on the fact that these festivals 3 This could relate to some possible ideas about a cyclical cosmic time (see schjødt 1981b, 1992, and also è39).
28 – Ritual Time and Time Reckoning
737
were highly correlated with the most intensive periods in the agricultural and pastoral working year. assuredly, snorri states that the festival at the beginning of sumar was as a sigrblót, emphasizing a martial aspect of this ritual. But this is clearly the warrior aristocrat’s perspective. for most people, spring was above all a time for sowing and animal breeding. little agricultural work was to be done in summer however, when the livestock grazed freely and the crops grew on the fields. The most intensive period in the working year took place during a period from late summer to early winter, which was the time of hay-making, harvest-ing, threshing, grinding, hunting, slaughtering of wild animals and livestock, brewing beer and mead, baking bread, desiccating fish and meat, as well as much other work to preserve food for the winter months. in some early medieval sources, it is said that much of this work was to be done before Christmas.
Concluding Remarks: The Ritual Year
The concept of time is closely related to religion. in old Norse mythology (cf. Vǫluspá 5–6), time was ordained by the gods in illo tempore to establish cosmic order from the primordial chaos. it was the gods who established the regulations of night and day, the cycles of the moon, and the daily movement of the sun in the sky. Hence, they even ordained the measurement of time in days, lunar months, and solar years to assist mortals in their daily lives. No wonder, then, that the germanic peoples could so easily equate the gods of the Roman day-names with their own emic deities, such as, for example, in the case of the scandinavian weekdays Týr’s day (Tuesday), Óðinn’s day (Wednesday), Þórr’s day (Thursday), and frigg’s day (friday).
since the cyclic movements and the astronomical fix points of the celestial bodies were fundamental natural manifestations of the ordered cosmos, as well as the foundations of the conception and the measurement of time, these phenomena also formed the bases for the ritual calendar. some late pieces of evidence suggest that sunrise and sunset might have been common times for certain pre-Christian daily rituals to be carried out, while varying sources dated from the first century ce to the late premodern era affirm that the rise of the new and full moon, as well as the solstices and equinoxes of the solar year, were important points in time for certain calendrical festivals.
summarizing the scant written evidence, it seems that the ritual year in ancient scandinavia consisted of the main calendrical festivals discussed above as well as listed below. some of these can be confirmed by archaeology. By determining the age of slaughtered young animals, it is sometimes possible to ascertain at what time of year a certain ritual took place (see also è27).
738
Andreas Nordberg
The West-Nordic dísablót seems to have been associated with a specific new moon in (approximately) october, which calendrically was related to the so-called ‘winter nights’ (old Norse vetrnǽtr, old swedish wintirnætir), representing the beginning of winter. Rituals during the autumn took place at frösö in Jämtland, as well as at skedemosse on Öland and at Käringsjön in Halland.
The pre-Christian yule festival at ‘mid-winter’ or ‘the mid-winter’s night’
(old Norse miðvetr, miðvetrarsnótt; old swedish midhvinter) seems to have coincided with the full moon of the lunar month (named after the festival) following the winter solstice.
another important festival related to the quarters of the year was held at the
‘the first day of summer’ or ‘the summer night(s)’ (old Norse sumarmál, old swedish första somardagher, somarnat) in mid-april. Rituals during the spring are attested at frösö in Jämtland and at Käringsjön in Halland.
according to some, albeit rather scant sources, public festivals were also held at ‘midsummer’ (old Norse miðsumar, old swedish miþsumar). Notably, major West Nordic legal assemblies such as the Norwegian Frostaþing and Gulaþing and the icelandic Alþing were likewise held in the middle of summer.
Most probably, these gatherings also encompassed religious activities. Rituals during the summer are known from frösö in Jämtland.
The major East scandinavian gathering Distingen in svetjud seems to have opened with the rise of the ‘Disting’s full moon’, that is, the full moon of a certain lunar month calendrically related to the time of the vernal equinox. This festival is indicated by written sources to have taken place at gamla Uppsala.
Moreover, some religious (and social/economic) gatherings seem to have been related to even longer astronomical cycles. several sources relate of festivals held ‘every nine years’ or annual festivals that were expanded ‘every nine years’. as this temporal description is inclusive (i.e., the last year in one cycle is also counted as the first year in the next cycle), it is in fact referring to a cycle of eight years, which is most likely to be identified with an eight-year cycle of the moon, known to have been used as a basis for the reckoning of time in accordance with a lunar-solar calendar in several societies.
Taken together, all this strongly indicates that the celebration of calendrical festivals was associated with the reckoning of time and even with the ritual recreation of time and the ordered cosmos.
29 – Cultic Leaders and Religious Specialists
olof sundqvist
almost all cultures have cultic leaders, often even different kinds of cultic and religious leaders, at an official level of society. some people are regarded as more skilful or useful than others when approaching the other World and performing rituals such as public sacrifices or divinations. There are also particular persons who preserve oral or written traditions. some appear as specialists in world-views or symbolic systems. one and the same individual may carry out several or all of these functions. in traditional societies, such specialization may develop to different degrees; some leaders appear as exclusive religious specialists, while others may assume other societal functions, too. in certain societies, the cultic leaders, particularly the religious specialists, may form an institutionalized, centralized, and hierarchic organization. This organization may monopolize certain cultic functions, and also normalize the official worldview, theology, dogmas, and ritual practice (cf. sabourin 1973).1 Religious differentiation, specialization, and organization are dependent on the contextual aspects and general structures of society: ‘as the scale and complexity of society increase and the division of labour develops, so too does the degree of religious specialization’ (Turner 2010: 144).
1 a process of bureaucratization may thus be discerned involving ‘rationality in decision making, relative impersonality in social relations, routinization of tasks, and a hierarchy of authority and function’ (Turner 2010: 145).
Olof Sundqvist, Professor of the History of Religions, stockholm University The Pre-Christian Religions of the North: History and Structures, Volume ii, ed. by Jens Peter schjødt, John lindow, and anders andrén, PCRN-HS 2 (Turn hout: Brepols, 2020) pp. 739–779
BREPols
PUBlisHERs
10.1484/M.PCRN-EB.5.116956
740
Olof Sundqvist
in what follows, cultic leaders and religious specialists will be presented chronologically and by means of a lexicographic (philological-terminological) approach. Thus, certain terms that may refer to religious or cultic functions will be in focus, such as: Proto-Nordic erilaz, gudija, * wīwaz (or * wiwaz), þewaz; the eighth-century term þulR (old Norse þulr); as well as old Norse goði, * ǫlvir,
*lytir, *vífill, gyðja, and vǫlva. Besides these religious officials there were also general leaders, such as chieftains (e.g., hersir sg.), jarls2 ( jarl sg.), and kings ( konungr sg.), whose offices included cultic or religious functions in society. These political leaders were involved in public sacrifices and rituals, and sometimes they appeared as organizers or commissioners of the religious feasts as well as custodians of the sanctuaries (e.g., sundqvist 2002, 2003a, 2003b, 2007, 2016). such
‘political leaders’ will not be treated here, since the relation between political leadership and religion is dealt with in (è23). Perhaps also the skalds should be included in the category religious or ritual specialists, since they made a kind of oral-poetic performances in the ceremonial hall (cf. Nygaard 2019).
Research Positions and Terminology
The issue of cultic leadership in PCRN has rarely been discussed in previous research. Besides surveys in handbooks on ancient germanic and scandinavian religion (e.g., de Vries 1956–57a: i, 393–406) and references in some philological and onomastic studies (e.g., andersson 1992b; Brink 1996a; Vikstrand 2001), only a few scholars have investigated ancient scandinavian cultic leaders thoroughly (e.g., Phillpotts 1912–13; Wesche 1937; Kuhn 1978: 231–42; sundqvist 1998, 2003a, 2003b, 2007). Previous discussion has been polarized; two lines of interpretation can be discerned, each represented by scholars from two different fields of study: namely, philology (onomastics) and the history of religions. some historians of religions, such as, for instance, folke ström, argue that the ancient scandinavians lacked a professional priesthood: according to him, there were no priests whose duties consisted exclusively of serving the deities.3 instead, the political leader, the king or chieftain, made contact with the deities at the public sanctuaries on behalf of the people at the sacrificial feasts 2 The meaning of the modern English cognate ‘earl’ has changed somewhat and to many conjures a very different picture, which is why we prefer jarl.
3 ström (1985: 72; 1983: 71). Cf. Hultgård (1997: 19–20); Näsström (2001: 76). see also Phillpotts (1912–13), Dumézil (1973c), and Ellis Davidson (1993). The archaeologist olaf olsen (1966: 55) has a similar point of view; see also Kuhn (1978). for the early discussion on this topic, see Phillpotts (1912–13: 264–65).
29 – Cultic Leaders and Religious Specialists
741
and in other rituals taking place at cultic sites. This type of religious leadership has often been associated with the notion of sacral kingship. Historians of religions who investigate the entire germanic area sometimes argue that priests existed in some areas. Jan de Vries (1956–57a: i, 401), for instance, states that in the southern germanic area, profane and sacred leadership functions were separated, whereas the situation in the northern area seems to have been different: Bei den südgermanen haben die Priester neben der weltlichen obrigkeit gestanden
[…]. Die Quellen, die wir für skandinavien besitzen, scheinen darauf hinzuweisen, daß die Trennung der weltlichen und priesterlichen funktionen hier nicht, oder jedenfalls nur sehr spät, stattgefunden hat.
(among the south-germanic peoples the priests stood beside the secular authorities
[…]. The sources that we have from scandinavia seem to point out that a separation of secular and priestly functions never, or at least only very late, took place there.) other scholars think that the scandinavians did have specialized priests. This opinion is represented mainly by philologists, especially specialists on onomastics. Klaus von see (1964) argues that the old Norse term goði refers to an exclusively priestly office. only in iceland, he believes, where the historical situation was very special, did the goði office develop into a leadership including several functions, such as law and other ‘secular’ aspects. He also states that originally, the germanic people distinguished strictly between religious and judicial aspects. This issue has been debated throughout the twentieth century and beyond.
John Kousgård sørensen (1989) states that a priestly class, in Danish præste-stand, existed in late iron age scandinavia. He focuses on the old scandinavian term - vé(r), -vi(r)/-væ(r) (Proto-Nordic * wīhaz), which he interprets as ‘priest’.
according to him, compound nouns including this term reflected a differentiated hierarchical priesthood. other specialists on toponomastics, such as lars Hellberg (1986a), Thorsten andersson (1992b), and lennart Elmevik (2003b), have observed terms in the placename material which could refer to exclusive religious specialists: for example, old Norse goði, *vífill, and *lytir.
in general, the specialists in toponomy apply designations such as ‘priests’,
‘priesthood’, or ‘pagan priests’ to ancient scandinavian conditions. one exception is Per Vikstrand (2001: 386, 427), who uses the concept kultfunktionärs-beteckningar (designations of cultic functionary).
The difference between historians of religions and philologists/specialists on toponomy is probably due to the different evaluation of the source categories and the different methodologies that the respective disciplines use. Historians
742
Olof Sundqvist
of religions rely mainly on historical and narrative sources, while philologists put more trust in the linguistic material and etymology (cf. Vikstrand 2001: 396). The difference of opinion is also caused by some confusion of terminology. it seems that philologists have used the common category ‘priest’ without any definition. Nor have historians of religions defined what they mean by
‘priest’. in this discipline, however, suggestions as to how operational concepts such as ‘priest’ and ‘priesthood’ can be conceived are sometimes encountered (see, e.g., sabourin 1973; Widengren 1969).
in an article published in 1998, olof sundqvist criticizes this terminology.
on the basis of classical phenomenological treatments, sundqvist proposes analytic definitions of the categories ‘priest’ and ‘priesthood’ and tests them on the scandinavian materials.4 He arrives at the conclusion that common features of priests or priesthood were vague in early scandinavia. There is only weak evidence for initiations into or formal training for a religious office.5 The cultic officials do not seem to constitute a closed and hierarchical form of organization that appears as an independent social stratum in society. Neither was there a priestly institution standardizing world-views or ritual practices. finally, the cultic leaders did not have or perform with external characteristics or imposed taboos, which would have helped to isolate them in society. one exception to this was the oath-ring used by the goðar (see below). Because of these circumstances and in consideration of the information provided by the literary sources that the rulers and chieftains on different levels of society were the people who performed important cultic functions at the sanctuaries, sundqvist argues that the concept ‘priest’ could be misleading in such treatments. since the concept of ‘priest’ (from greek presbyteros ‘the older’, presbys ‘old person’) was, moreover, formed and developed within a Christian tradition, sundqvist later argues that it is better to use more neutral concepts in germanic and scandinavian contexts, such as cultic leaders, cultic performers, or religious specialists, in order to avoid serious misinterpretations (sundqvist 2003a, 2007; cf. Hewitt 1996: 16; Rüpke 1996: 241). This type of criticism has its limitation, since the concept of a ‘priest’ could be used as an etic construction and an operational concept, completely defined by the analyst (cf. sinding Jensen 2014: 7) and thus more or less freed from its ordinary emic use and associations in, for instance, Christian contexts.
4 for definitions of ‘priest’ and ‘priesthood’, see, e.g., sabourin (1973), Widengren (1969), and sundqvist (1998).
5 Traces of initiation may be seen in mythical traditions, however; see mainly schjødt (2008). Cf. sundqvist (2009b, 2010).
29 – Cultic Leaders and Religious Specialists
743
in the present chapter, the description ‘cultic leader’ is proposed.6 it refers here to a person who was temporarily responsible for certain religious functions in society at different types of cultic sites, both communal sanctuaries and those located at the ruler’s farm (see below). The cultic leader had other societal duties beside his or her religious tasks. He or she also functioned as a general political leader. The term ‘religious specialist’ designates an exclusive religious office in the present overview, that is, it describes that a more intensified and permanent specialization has taken place and that the religious leader has become more or less professional (cf. Rüpke 1996; Turner 2010). it is, however, almost impossible to say whether we are dealing with cultic leaders or religious specialists in the Continental, anglo-saxon, and early scandinavian materials (100 bce–800 ce), due to the scant and fragmentary sources. When treating Viking age conditions (800–1100), we may at least formulate hypotheses regarding this matter.
Continental and Anglo-Saxon Sources ( c. 100 bce to 1000 ce) When the Romans described foreign peoples’ beliefs, rites, and customs, they occasionally used indigenous concepts but often they applied latin terms, such as sacerdos ‘priest’. it is hard to know what lies behind such latinizations. often, it may concern an interpretatio Romana or an ethnographic cliché. Moreover, when the Romans described germanic cultic leaders and religious specialists, they rarely produced a coherent image. Caesar stated during the first century bce that the southern germanic peoples had no druids ( druides) who served the gods of the cult ( De bello Gallico 6.21). in contrast to that, Tacitus mentions
‘priests’ ( sacerdotes) among the northern germanic tribes in his Germania. in the Nerthus cult, which existed among seven tribes somewhere close to the Baltic sea (perhaps in Denmark), a ‘priest’ ( sacerdos) had a special relationship to the goddess and her shrine ( Germania ch. 40). a ‘priest’ ( sacerdos) also performed ritual roles among the Naharvali ( Nahanarvali), a tribe belonging to the religiously based covenant called Lugii. They lived in Central Europe, north of the sudetes mountains in the basin of the upper oder and Vistula Rivers, that is, in the south and middle of modern Poland. Tacitus says that 6 The concept ‘cultic leaders’ and eqvivalents to this term, such as ‘cultic/ritual/religious specialists’, swedish/Norwegian kultledare/ kultledere or kultfunktionär/ kultfunksjonær, have been applied by several scholars working with ancient scandinavian and germanic contexts.
see, e.g., Hultgård (1997), sundqvist (1998, 2003a, 2003b, 2007, 2016), Vikstrand (2001), steinsland (2005a, 2007), Näsström (2001), Tausend (2009), and gardela (2012).
744
Olof Sundqvist
their ‘priest’ ( sacerdos) presided in a particular sacred grove, wearing female dress ( muliebri ornatu), and that in this grove, the divine twins called alcis/
alci were worshipped ( Germania ch. 43). in addition to the important functions in divination rites ( Germania ch. 10), the germanic ‘priest’ also had legal functions, according to Tacitus. only he had the right to punish a warrior with death or to distribute other forms of punishment ( Germania ch. 7, 11).
The conflicting data of Caesar and Tacitus may be because one of them or both were poorly informed about cultic conditions among the germanic peoples, or it may stem from the fact that they described various germanic tribes at different times. There are about 150 years between Caesar’s and Tacitus’s works. Caesar observed the germanic tribes in connection with warfare.
Perhaps the Roman emperor wanted to underline the differences between the fascinating role of the druids in the Celtic religion and the more underdevel-oped germanic cult, in a more or less rhetorical way (cf. de Vries 1956–57a: i, 397–99, Turville-Petre 1964: 261; Kuhn 1978; Timpe 1992; Polomé 1992).
it is hard to know what Tacitus actually meant by sacerdos and sacerdotes. it has been suggested that his expression sacerdos civitatis ( Germania ch. 10) referred to a more specialized and limited cultic leadership (de Vries 1956–57a: i, 398).
However, one cannot discern any sharp specialization, differentiation, or hierarchy within the group of sacerdotes that he describes. However, Tacitus makes a particular distinction between sacerdos civitatis and rex vel princeps civitas ( Germania ch. 10), suggesting that the germanic peoples differentiated between religious and secular leadership. Both seem, however, to be equally concerned with divination rites. The ruler, according to Tacitus, also fulfils ritual functions and participates in the same divination rites as the ‘priest’. furthermore, Tacitus distinguishes between public worship and private worship of which the latter was led by patres familiarum. The religious functions of the princeps, sacerdotes civitatum, and patres familiarum, then, seem to be only one aspect of their more general social (and political) leadership roles (cf. Timpe 1992: 484).
The question of whether the germanic cultic leadership was hierarchically organized has likewise been discussed (cf. de Vries 1956–57a: i, 398–99).
ammianus Marcellinus (fourth century ce) mentions that the Burgundians had a sacerdos omnium maximus, ‘high-priest’ (28.5.14). The title might reflect an official position at the top of a hierarchy. it is supported by a statement from the anglo-saxon cleric Bede. He called the official Coifi primus pontificum, that is, one who possessed the highest priestly ministry ( Historia Ecclesiastica 2.13).7
7 Researchers have, however, revealed traces of foreign or later ideas in these sources and names (e.g., Kuhn 1978).
29 – Cultic Leaders and Religious Specialists
745
There are native germanic designations in the gothic Bible referring to either cultic leaders or religious specialists. The term ( auhumists) weiha appears as a translation of the greek term archiereus (‘archpriest’, ‘high-priest’;
‘oberpriester’, ‘Hoherpriester’) (John 18.13), which may also support a hierarchical organization. Wulfila also uses the gothic term gudja when designating the Jewish ‘priest’ (greek hiereus ‘priest’, cf. sa auhumista gudja for archiereus
‘high priest’). it is derived from the gothic noun guþ ‘god’. There is no doubt that gudja (a jan-stem) is an old word that corresponds to Proto-Nordic gudija and is related to old Norse goði (an an-stem). The word is probably also related to old High german gotinc/ goting, used to gloss tribunus ‘chief, commander’
(cf. Wesche 1937: 6–8; de Vries 1956–57a: i, 400, 1962a: 181; green 1998: 33–34; critically reviewed by Kuhn 1978: 235–36). it has been argued that this word should be associated with an old High german *goto (Wesche 1937: 6).
in a lost manuscript, a derivation of it — namely, the old High german verb gotten — is glossed iustificare ‘act justly towards, do justice to, vindicate, justify’
(Wesche 1937: 6–7; green 1998: 33–34. Critically reviewed by Kuhn 1978).
Whether the terms gudja, goði, and gotinc originally refer to individuals who were regarded as exclusively religious specialists is therefore uncertain.
There are West germanic glosses from the post-Roman period that can refer to cultic leaders or religious specialists. The old High german words bluostrari, harugari, and parawari can be regarded as later word formations based on the latin loan suffix -ari- (Wesche 1937: 2–6; de Vries 1956–57a: i, 399; green 1998: 28–29).8 Wulfila ( John 9.31) related the expression theosebes
‘anyone who is devout’ to gudblostreis ‘worshippers of god’, a compound where the second element is derived from the gothic verb blotan ‘to worship’ (cf. guþ blotan for theosebeia ‘religion, piety, the service of god’ and old Norse blóta ‘to worship, sacrifice’). The old High german term bluostrari and gothic blostreis can be linked to other native germanic terms, such as old Norse blótmaðr and old English blōtere ‘sacrificer’ (cf. Wesche 1937: 5–6). The latter word formations seem basically to be constructed from native terminology. The term harugari, linked to old High german harug, old English hearg, old Norse hǫrgr,
‘cultic place’, has been interpreted as ‘priest’, ‘diviner’; while parawari, from old High german baro ‘sacred forest (grove)’, can be interpreted as ‘guardian of the sacred forest (grove)’ (Wesche 1937: 2–6; de Vries 1956–57a: i, 376, 399; green 1998: 29).
Perhaps the most common old High german and Middle High german
word for a Jewish or Christian ‘priest’ is ēwart(o), ēwart. This is equiva-8 it should be noticed, however, that this suffix appears as early as in gothic.
746
Olof Sundqvist
lent to old English ǣweweard, which literally means ‘guardian of the law’, but which may also have had the meaning ‘priest’ (Wesche 1937: 8–16; de Vries 1956–57a: i, 399, Kuhn 1978: 234; green 1998: 31–33). The word is derived from the old High german ēwa, ēa (cf. old High german, Middle High german ē, ēwe; old saxon eo; old frisian ā, ē, ēwa, ēwe; old English ǣ, æw), which may be translated as ‘secular law’, ‘law’ but also as ‘divine law’, ‘religion’ (Wesche 1937: 13, 20; green 1998: 31). old High german ēsago, old English ēosago, and old frisian āsega means ‘law-speaker’, ‘judge’ but is also related to ēwart(o), ēwart (Wesche 1937: 12–19; de Vries 1956–57a: i, 399).
old frisian āsega occurs as a translation of sacerdos in Christian contexts.
Whether these terms indicate that legal aspects and religious cult were linked together as one office in the pre-Christian context is controversial (cf. von see 1964: 92–93, 105–12).
There are few indigenous designations for cultic leaders and/or religious specialists among the anglo-saxons. The old English term ǣweweard has already been mentioned. King alfred’s English translation ( c. ninth century) of Bede’s Historia (2.13) ( c. eighth century), describes primus pontificum Coifi in Northumberland as ealdor-bisceop. He belonged to the group called ealdor-men
‘the eldest’. Coifi, who lived in the seventh century, was associated with a shrine that he desecrated when converting to Christianity. according to Bede, the old cultic leaders could neither carry weapons nor ride stallions in the vicinity of the shrine. it is likely that the episode in Bede’s book, written down more than a hundred years after the events described, was influenced by Christian attitudes to pagan cultic leadership (Kuhn 1978)
in older sources from the Continent, female cultic leaders or perhaps rather religious specialists also appear.9 already strabo (7.2.3) mentions that certain women of the Cimbri (perhaps from Jylland) had important functions within 9 Rudolf simek (2015) states that concepts such as ‘cultic leaders’ or ‘religious specialists’
are not suitable for female ‘cult functionaries’ and ‘sorceresses’ appearing in these sources. He admits that these women may have had ‘connections with the supernatural’ and ‘prophecy’; however, since they had no associations with a ‘public cult’, they should rather be described as
‘prognostic specialists’. simek’s interpretation of ‘cult’ in this context seems to be very narrow, since it excludes more private cultic and ritual activities such as divinations and sacrifices at farms (cf. Ringgren 1970: 89–93 and lang 1993, who give the concept of cult a much wider frame). in the Viking age scandinavian context (i.e., in the old Norse sources), we meet the vǫlur, for instance, who performed divination rituals at individual farms (see below). in our opinion they should definitely be regarded as ‘religious specialists’, although they sometimes performed rituals on a more private level (see below). some female ‘sorceresses’ in the early Continental sources could be likewise labelled.
29 – Cultic Leaders and Religious Specialists
747
sacrifice and divination rites. These ‘priestesses’ (pl. hiereiai), who wore special clothes, received prisoners of war with swords in their hands and crowned them with garlands. Then they made them bend down over a cauldron and cut their throats. Through the blood that ran down into the cauldron, the ‘priestesses’
could set a prophecy ( manteia), and by means of observing the intestines of the prisoners, they could make prophecies of victories for the people (è25). The credibility of this story has been questioned (de Vries 1956–57a: i, 393; simek 2015). it may be a ‘horror story’ circulating among Roman soldiers (Zanker 1939); however, the ritual process described is consistent with information on Celtic practices and customs attested in other written sources (e.g., Diodorus 5.31.4) and in archaeological finds from Duchcov and Vix (Bourriot 1965; Ellis Davidson 1993; Hultgård 2002a).
Caesar tells of matres familiae among the Suebi, who by means of lots and divination rites could determine whether it was appropriate to engage in battle ( De bello Gallico 1.50; cf. Polomé 1992). also Tacitus mentions that germanic women had something holy and prophetic about them (‘etiam sanctum aliquid et providum putant’) and that the men did not reject their advice ( Germania ch. 8). in this passage, Tacitus also mentions a woman called Veleda. she was worshipped as a goddess during Emperor Vespasian’s time. Earlier, also albruna (?) (in the manuscripts the name is variously rendered Auriniam, aurimam, aurinam and Albriniam) and other women were objects of some cult (cf. Helm 1913: 285; de Vries 1956–57a: i, 321; critically reviewed by Peterson 2002). Veleda is also mentioned in Tacitus’s Histories (4.60–62; 5.22–25), where it is said that she had a significant role during the Batavian rebellion in 69 ce. it is said that she inhabited a high tower from which she reported oracles to the people. Her name has been associated with Early irish fili(d) ‘poet, scholar’ (Dichter, gelehrter), suggesting Celtic influences (de Vries 1956–57a: i, 320, 404; guyonvarc’h 1961; simek 2007, 2015; critically reviewed by Meid 1964). Cassius Dio (67.5) mentions a female called ganna among the Semnones, in the Svabi-tribe, who succeeded Veleda.
Her name may be related to old Norse gandr ‘magic staff ’, alluding perhaps to her magical abilities (de Vries 1956–57a: i, 297–98, 321, 324 and ii, 362; simek 2015: 74).
in the late eighth century ce, Paul the Deacon (1.3.7–9) mentions a gambara who turned to frea when her people were involved in a battle (cf. Origo Gentis Langobardorum 1; see also Gambaruc in saxo’s Gesta Danorum 8.13.1). Her name has been associated with *gand-bera ‘staff carrier’ (Helm 1946: 22, n. 49; ström 1985: 85, 272).
748
Olof Sundqvist
figure 29.1. Runic amulet of bone from
lindholmen in skåne, dated to about 400–550 (DR 261, Samnordisk runtextdatabas, lunds Universitets Historiska Museum no. 5084). The inscription includes the term erilaz.
Photo: Roberto fortuna, Nationalmuseet, Copenhagen.
Early Sources from Scandinavia ( c . 200–800 ce)
There are several possible native designations of cultic leaders and/or religious specialists in the oldest runic inscriptions from scandinavia: namely the terms erilaz, gudija, and perhaps the term *wīwaz (or *wiwaz; cf. *wīwilaz), which appears as a proper name ( Wīwaz/ Wiwaz) in the runic stone of Tune, Norway.
also early Viking age þulR and Proto-Nordic þewaz could, in certain cases, perhaps be added to them. Most likely, these titles reflect cultic leaders or religious specialists. Needless to say, the fragmentary information from the Proto-Nordic runic inscriptions can only lead to hypothetical reasoning.
Proto-Nordic erilaz** **
The Proto-Nordic term erilaz ( irilaz) is attested in at least eight runic inscriptions, all dating from the Migration Period.10 These inscriptions appear in different contexts, such as runic stones, gold bracteates, a fibula, a bone amulet, and a deposited spear-shaft (Krause and Jankuhn 1966; Düwel 1992b, 2015; Nowak 2003; Düwel and Nowak 2011). The word erilaz has been interpreted as ‘priest, magician, wizard’ or ‘rune-master’ (see survey in Düwel 1992b and 2015). some scholars have argued that the concept erilaz later developed into the old Norse concept jarl (Hellquist 1957; Klingenberg 1973). There are, however, linguistic problems with such an interpretation (andersen 1948; cf. Mees 2003). others have tried to relate it to the designation of a people called Heruli (latin eruli, heruli; greek erouloi) (Elgqvist 1952; cf. Hellquist 10 There is also a bracteate from Trollhättan with an inscription, which may be interpreted Ek erilaz Mariþeubaz haite, wrait alaþo(?) (i the eril am called Mariþeubaz (= the sea thief /
the famous thief ), i wrote a nourishing charm(?)) (Vg iK639, Samnordisk runtextdatabas). Cf.
axboe and Källström 2013. Perhaps there is a tenth erilaz-inscription on a fragment on a stone from strängnäs, found in 1962 (sö fv2011;307 U, Samnordisk runtextdatabas) which could be interpreted as … [ e] rilaz Wōdinz. see Hultgård (2010a) and gustavson (2011).
29 – Cultic Leaders and Religious Specialists
749
1957; Mees 2003; spurkland 2005). Recently, it has been suggested that erilaz was a leader of a type of secret and bellicose group (fischer 2005; è24) and that it designated a military function (Herschend 2005). since erilaz appears with other personal names, it has been suggested that this term is a title or refers to an office of some sort (Düwel 1992b, 2015; Hultgård 1998b).
There are some features of the runic inscriptions suggesting that the term erilaz could refer to a cultic leader or a religious specialist. some of them follow a pattern that indicates a ritual or cultic formula (Hultgård 1982, 1998b; sundqvist 2009a.; discussed by Düwel and Nowak 2011). immediately before erilaz there is often an emphatic ec ‘i’. after that, we may see the verb *haitan (cf. gothic haitan) ‘be called’ and a personal name. sometimes, there is also a mention of what erilaz does or what function he has. This pattern is visible in, for instance, the inscriptions of Kragehul (DR 196, Samnordisk runtextdatabas), lindholmen (DR 261, Samnordisk runtextdatabas), and Järsberg (VR 1, Samnordisk runtextdatabas). Krause and Jankuhn (1966) interpreted these inscriptions thus:
Kragehul:** ek(eril(aRasugisalas±mu(ha(haite(ga(ga(gaginu(ga(he /// lija /// hagalawijubig /// **
ek erilaR A(n)sugīslas mūha (oder Mūha) haitē. (ga(ga (ga (= gibu auja oder gebu ansuR), ginu-(ga . he[lma-tā]lija (oder: -[tā]lija[tō]) hagla wī(g)ju (oder: wī(h)ju)
bi g[aiRa] ///
ich Eril (= Runenmagiker) heiße Ásgísls gefolgsmann (oder: sohn Muha). ich gebe glück (oder: gabe-ase) (dreimal), magisch-wirkendes (Zeichen) (ga. —
Helm vernichtenden (?) Hagel (= Verderben) weihe ich an den speer.
(i the Eril (rune-master), the vassal (or: son) of Ásgísl, am called Muha. i give luck (or:giving-áss) (thrice), [magical sign] (ga. i dedicate helmet-destroying hail (=
destruction) by means of the spear.)
lindholmen: (a) ek erilaR sā wīlagaR ha(i)teka: (B) aaaaaaaa RRR nnn x b m u ttt: alu ich, der Eril (= Runenmagiker) hier heiße ‘listig’. aaaaaaaa RRR nnn x b m u ttt: alu [= Zauber].
(i, the Eril, am called the Cunning one. [cryptographic sequence] + alu [= magic].)11
11 Krause and Jankuhn (1966) do not present any transliteration here. Samnordisk runtextdatabas reads it:** ek erilaz sa wilagaz hateka: aaaaaaaazzznnn-bmuttt: alu:**
750
Olof Sundqvist
Järsberg: ubaRhite: (haraban(aR (hait ek(erilaR runoRw aritu ŪbaR(?) h(a)itē, HrabnaR hait(ē); ek erilaR rūnōR wrītu.
Der Tückische (oder …ub) heiße ich, Hrabn heiße ich; ich Eril (= der Runenmeister) ritze die Runen.
(The Wicked (or … ub) i am called, Hrabn i am called; i Eril (rune-master) carve the runes.)12
a similar formula ( ek + *haitan + name + exploit/function) is also attested in the inscription of two identical C-bracteates from Køge (iK 98 sjælland ii, Die Goldbrakteaten), which do not contain the designation erilaz: hariuhahaitika: **farauisa: gibuauja: )ttt **(?). ** **Krause and Jankuhn have translated this sequence as: ‘Hariuha heiße ich, der gefährliches Wissende. ich gebe Heil’ (Hariuha i am called, the one knowing dangerous things, i give luck).13 an emphatic ek
‘i’, a designation of a function, and a description of the subject is also visible in the Proto-Nordic inscription from Nordhuglo, Norway (see the text below). in this inscription, the verb *haitan has been left out. The functional designation erilaz has further been substituted with the designation of the cultic leader gudija. The self-predication and formulaic language indicate, however, that this inscription has a connection to the erilaz- inscriptions.
These statements including the formula haiteka that appear in the Proto-Nordic runic inscriptions have an interesting relation to some myths and rituals associated with the god Óðinn (e.g., Hultgård 1982, 1998b, 2007b, 2010a).
They recall some phrases found in the Eddic poem Grímnismál st. 46–50, where Óðinn introduces himself using the expression hétom ec, for instance in st. 46: ‘Hétomc grímr, | hétomc gangleri’ (i was called Mask, | i was called Wanderer). This presentation ends in stanza 54, where Óðinn makes a contrast between the name he carries now and the names he has previously had: ‘Óðinn ec nú heiti, | yggr ec áðan hét | hétomc Þundr fyrir þat’ (Óðinn i am called now, | Terrible one i was called before, | they called me Thund before that).
12 The runologist Magnus Källström states that Erik Moltke’s (1981) reading and interpretation is better than those of Krause and Samnordisk runtextdatabas: ek(erilaR / ubaRh[a]ite: (haraban(aR / (hait[e] / runoR w / arit / u (i the eril (a kind of domestic tutor or goði) am called the adversary (properly: he who stands in hostile opposition), my name is raven, i write (the) runes (the inscription)). see axboe and Källström 2013.
13 Krause and Jankuhn (1966). **farauisa **can be interpreted as fārawīsa ‘the one knowled-gable of danger’, or farawīsa ‘the one who knows about journeys’. see Düwel (2005a). for some other problems with the interpretation of this inscription, see Düwel and Nowak (2011).
29 – Cultic Leaders and Religious Specialists
751
These formulistic expressions with an emphatic ec ‘i’ may very well be reminiscent of cultic self-predications or prayers. They were most likely recited in ritual contexts, for instance, at the sacrificial feast or when invoking the deity.
it is possible that the cultic leader/religious specialist in such rituals was seen as a representative of or deputy for the god. similar ritual formulae, so-called aretalogies, may be seen in late antique cults from greece and Rome and also in cults from ancient iran (feist 1922; Hultgård 1982; 1998b, 2010a; sundqvist 2007: 193–218).
Perhaps the erilaz-inscriptions could also be interpreted as such divine self-predications. The bynames used by the erilaz sometimes have connections to the names, functions, or attributes of Wodan-Óðinn. This relation has been noticed by several scholars previously.14 on the Järsberg-stone, for instance, this sequence appears: (haraban(az (VR 1, Samnordisk runtextdatabas). Both Krause and Jankuhn (1966: 157–58) and Düwel (2008: 36) interpret it as a byname of erilaz, Hrabnaz. They argue that this name is based on a Proto-Nordic * hrabnaz (cf. old Norse hrafn) ‘raven’. Óðinn’s relation to ravens is well known from the old Norse traditions (see Mitchell forthcoming). sometimes, he carries names or bynames which refer to them. in the old skaldic poem Haustlǫng st. 4 ( c. 900), by Þjóðólfr ór Hvini, he is called Hrafnáss ‘the raven-god’ (cf. hrafna guð ‘the god of the ravens’ in Gylfaginning p. 38),15 and in Húsdrápa st. 10 ( c. 990) he is designated Hrafnfreistuðr ‘raven-tester’. The latter name refers most likely to the event when Óðinn tested his ravens by sending them out in the world in order to collect information (cf. old Norse freista
‘to try, to put to a test’) (falk 1924: 18). in a lausavísa composed by Hallfreðr vandræðaskáld in the late tenth century, Óðinn is called hrafnblóts goði, ‘the goði of the raven-sacrifice’ (see falk 1924: 18 and sundqvist 2009a; critically considered by Wulf 1994).
it should be noted, however, that not all names to which the erilaz references relate in the runic inscriptions may be attached to this specific deity (e.g., Elmevik 1999; stoklund 2001b; sundqvist 2009a and the literature therein).
Nevertheless, it seems that erilaz also appears in other cultic contexts which may be linked to Wodan-Óðinn. according to the inscription on the spear-shaft from Kragehul, the erilaz presumably consecrates the sacrificial gift: ‘i erilaz … con-14 for example, Marstrander (1952); Krause and Jankuhn (1966); Müller (1975); Hultgård (1982, 1998b); Dillmann (2003b); sundqvist (2009a). Critically considered by Wulf (1994) and Nowak (2003).
15 Hrafnáss also appears in the Poem about Gizurr gullbrárskáld by Hofgarða-Refr gestsson (eleventh century).
752
Olof Sundqvist
secrate (or hallow) ( wīju)’ (see above; cf. DR 196, Samnordisk runtextdatabas).
The wooden spear was broken into at least five pieces when it was found. This makes the reading and interpretation of the inscription uncertain. However, some runologists and historians of religion accept that the verb * wīhian ‘consecrate, hallow’ is part of it (e.g. Düwel 2008, 2015; Hultgård 2010a).16 The spear was deposited in a bog where other sacrificial finds have also been detected (ilkjær 2001; stoklund 2001a; Hultgård 2010a). The archaeological evidence thus indicates that the entire spear was sacrificed. in the old Norse sources, Óðinn is often related to sacrifices or consecrations which involve spears.17 We therefore cannot exclude that erilaz in Kragehul consecrated/dedicated both the spear and the enemies, who were killed by it, to Wodan-Óðinn, that is, the spear-god among the æsir (McKinnell and others 2004).
another argument supporting the notion that persons called erilaz were associated with religious-magic matters is the fact that the inscriptions containing this word often appear on objects that can be regarded as ritual or sacrificial, such as the spear-shaft from Kragehul. The word erilaz and the formula that apparently goes with this term also appear on the object made of bone or horn from lindholmen (DR 261, Samnordisk runtextdatabas), which has been interpreted as an amulet (e.g., stoklund 2001b; McKinnell and others 2004; Hultgård 2010a). Erilaz also turns up on the two identical f-bracteates from Äskatorp and Väsby (iK 241.1 and iK 242.2, Die Goldbrakteaten). They display an image that may relate to the mythical sphere. The name Wīgaz (* wīgaz
‘fighter’), which appears in this inscription, can also be associated with a heiti of Wodan-Óðinn (Düwel and Nowak 2011: 428).
it should be mentioned that erilaz could also be linked to other societal functions besides cultic assignments. it seems as if the erilaz also handled runes.
in the Järsberg inscription, for instance, we read: ‘ek erilaz … runoz writu’ (i erilaz … carve runes) (see above; cf. VR 1, Samnordisk runtextdatabas), and on the bracteates from Väsby och Äskatorp, similar information appears: ‘i stained ( fāhidō) this piece of art, Wīgaz, i erilaz’. The handling of runes may of course have been one aspect of a cultic office, since runes sometimes appear to have a religious dimension. according to the passage called Rúnatalsþáttr Óðins in Hávamál (st. 138–45), for instance, the runes are linked to Óðinn (sundqvist 2009a, 2009b, 2010; è42) and in stanza 80 it is furthermore stated that: ‘Þat 16 it could also be related to the verb * wīgian (old Norse viga, vega) ‘combat’. Cf. stoklund (2001a).
17 see Vǫluspá st. 24; Hlǫðskviða st. 27–28; Ynglinga saga ch. 9; Gautreks saga ch. 6–7. see further sundqvist (2009b, 2010).
29 – Cultic Leaders and Religious Specialists
753
er þá reynt, er þú at rúnom spyrr, inom reginkunnom’ (it is then tested when you ask about the runes derived from the gods).18 But perhaps we may also see the erilaz as a leader who had several societal functions, both religious and profane, and thus could be considered a cultic leader.
Proto-Nordic gudija** **
as noticed above, the formula connected to the erilaz-inscriptions also has a close parallel in the Proto-Nordic runic inscription from Nordhuglo, Hordaland in Norway, dated to the fifth century. in this latter inscription, we may find an emphatic ec ‘i’ just before the designation of an office or societal function followed by a description of the subject. Krause and Jankuhn (1966) interpret it as: ekgudijaungandiRih /// ( ek gudija ungandiR ih …) ‘i, the priest, immune to sorcery, (or: who does not engage in sorcery), in Huglo’. This expression may also be some kind of self-predication, applied within a cultic context.
The Proto-Nordic word gudija corresponds to the gothic gudja ‘priest’, ‘sacrificer’ and is closely related to old Norse goði ‘chieftain’ or ‘cultic leader’ (see above and below). These terms are derived from a word with the basic meaning
‘god’. The word ungandiz in the Nordhuglo inscription may be taken to confirm that the gudija deals with religious-magic aspects. The term must be related to old Norse gandr ‘sorcery’, which refers to different types of witchcraft. The privative prefix un- indicates that the word can be interpreted either as ‘he who is immune to sorcery’ or ‘he who does not engage in sorcery’ (spurkland 2005: 48–49; Düwel 2002). Perhaps the context of the Nordhuglo inscription could also support a cultic interpretation. The inscription was found on a stone near the farmyard, nearly 3 m high and 70 cm wide. it has been suggested that the stone originally marked a grave mound near the sea. Possibly, the gudija at Huglo was somehow involved in making the grave monument.
Perhaps the gudija at Huglo was also involved in other societal functions, just like the icelandic goði (see below). in that case, he should be regarded as a cultic leader.
18 This could be compared to the expression on the Noleby (sixth century) inscription: runo fahi raginakudo (i paint a rune, derived from the gods) (Vg 63, Samnordisk runtextdatabas). Cf. Düwel (2008: 35).
754
Olof Sundqvist
Proto-Nordic *wīwaz** (or *wiwaz****)** Traces of a cultic leader or religious specialist may also occur in the first part of the Tune inscription (fifth century), østfold, Norway (N KJ72, Samnordisk runtextdatabas). James Knirk (2006) interpreted this sequence thus: ekwiwazafter · woduri / dewita(da(halaiban: worahto:? …????
KJ56 † U, Samnordisk runtextdatabas). gunter Müller (1968) is inspired by Krause and Jankuhn in his interpretation of the names Wīwaz and Wīwila, but he has a different opinion about the word formation. Wīwaz should be related to the word * wīgwaz, which was construed directly from a verb related to old Norse vígja. it should be interpreted as ‘inaugurated, consecrated’. linked to
* wīgwaz, there was a diminutive construction * wīgwilaz with the meaning ‘the small consecrated one’. Müller argues that the old Norse name Vífill is derived from the diminutive form (see below). This interpretation is, however, somewhat uncertain (see, e.g., Peterson 2007; cf. Vikstrand 2009a).
ottar grønvik (1987b, cf. 1981) argues that the name on the Tune stone should be read with a short vowel in the root syllable, *Wiwaz, and be interpreted as the one who consecrates (sanctifies, blesses) and denominates a kind of cultic leader. He argues that this name was visible also in other inscriptions, such as the one on the Eikeland clasp (N KJ17a $U, Samnordisk runtextdatabas). There, he found the sequence wiz, which he interprets as *Wiwaz. He likewise interprets the sequence wiwio on the Eikeland clasp as a female derivation of *Wiwaz. since the subject of the inscription wiz describes wiwio (Proto-Nordic *Wiwjō, a genitive form of *Wiwja) as his asni, that is, the dative of an unattested noun asniz ‘beloved’ (cf. old Norse ást ‘love’), he suggests that
* Wiwaz and *Wiwja were a married couple (critically considered by Knirk 2015). John Kousgård sørensen (1989) opposes grønvik’s interpretation and 19 Cf. Marold (2015). Knirk (2015: 431) states that this interpretation is uncertain because * Wiwaz could etymologically speaking just as well be related to ‘fight’ as to ‘holy’.
29 – Cultic Leaders and Religious Specialists
755
argues that wiz on the Eikeland clasp should be interpreted as * Wīwaz, that is, a substantivization of an adjective * wīha- ‘holy’ and with the meaning ‘priest’.
The name on the Tune stone, wiwaz, is interpreted as an original * wīha-wīhaz
‘præst ved vi’, that is, a ‘religious specialist at a vi-sanctuary’. also this interpretation has been regarded as problematic (e.g., Peterson 1994; Knirk 2015).
Despite this uncertainty, several runologists relate the appellative behind the name Wīwaz/Wiwaz ( *wīwaz/*wiwaz) in the Tune inscription to some kind of cultic leader or religious specialist, and perhaps also the name Wōdurīdaz should be included. it may correspond to an old Norse *Óð-ríðr, where the first element is equivalent to the old Norse adjective óðr ‘frenzied, raging, raving: mad with fear, angry, insane, violent’ and the second element is ríðr ‘rider’. The entire name could thus possibly be interpreted as a Wodan-Óðinn- heiti ‘the frenzied (violent) rider’ (cf. spurkland 2005). Wōdurīdaz was most likely also a local ruler, since he is called: wita(da(halaiban ‘he who provides (takes care of ) bread’, ‘he who secures bread’ (cf. old English hlāf-weared, hlāford ‘lord’) (Marstrander 1930; Brink 2010a; Marold 2015).
We cannot rule out that there was an appellative *wīwaz/ *wiwaz designating some type of religious leadership during the Roman and Migration Periods.
it is, however, impossible to say exactly what kind of ritual functions he had or whether he should be classified as a cultic leader or a religious specialist.
The Eighth-Century Term þulR** **
in old Norse texts, the term þulr is used with reference to a ‘wise man’, ‘sage’,
‘cult orator, pagan priest’ (cf. old English þyle) (see, e.g., olrik 1909; Vogt 1927, 1942; simek 2007; Brink 1997; sundqvist 1998, 2003a, 2003b, 2007; Poole 2005; and most recently Tsitsiklis 2017). This noun is often related to the old Norse verb þylja ‘speak, mumble, sing’. There is a certain genre of poetry called a þula, that is, a rhymed or alliterative list or chant. The equivalent runic word þulR appears in the Danish snoldelev inscription (DR 248) at sylshøj, sjælland: kun’uAlts| |stAin ‘ sunaR ‘ ruHalts ‘ þulaR ‘ o salHauku(m) ‘gunvald’s stone, son of Roald, þulR in salløv’ (see Moltke 1985; cf. Düwel 1992b, 2008, 2015; stoklund 2005). The last sequence of the inscription is literally interpreted as: a salhaugum ‘on the sal-mounds’. This indicates a connection between the þulR and the hall building, that is, the sal, wherein ceremonial feasts were performed (Brink 1996a). Perhaps the þulR performed ritual utterances during feasts in the hall. This inscription is dated to 650–800/900 (Moltke 1985).
The stone was part of a monument containing fifteen stones arranged in two lines and located close to the mound called Blothøj (see Danmarks runeind-
756
Olof Sundqvist
figure 29.2. The rune stone at
snoldelev in sjælland (DR 248,
Samnordisk runtextdatabas).
The inscription includes the term
þulz. Photo: Roberto fortuna,
Nationalmuseet, Copenhagen.
skrifter). it has interesting symbols. The swastika and the triskele of three drinking horns are coeval with the inscription, although the ‘sun wheel’ is prehistoric (see Danmarks runeindskrifter). The stone contains two lines of runes, written mainly with the twenty-four character ‘elder futhark’ (Moltke 1985). This inscription therefore provides authentic evidence of a þulR during the Vendel Period or early Viking age in southern scandinavia.
it has been discussed what kind of functions the þulR/ þulr had. gun Widmark suggests that a þulR was the keeper and intermediary of the Runic swedish mogminni ‘memory of the kin (people)’ mentioned in the Rök inscription, Östergötland, sweden (Ög 136, Samnordisk runtextdatabas) (Widmark
29 – Cultic Leaders and Religious Specialists
757
1993, 1997; gustavson 2003; Ralph 2007; Holmberg 2015; see also Holmberg and others forthcoming). such memories were ritually declaimed at certain occasions, for example, the burial feasts called erfi. it has also been argued that, in the germanic area, the succeeding ruler, often a younger or youngest son, was required in connection with the inheritance feast to learn numinous esoteric knowledge about runes, royal genealogies, and spells in order to win power and attain the office of ‘sacral kingship’ (fleck 1970; sundqvist 2002; schjødt 2008). according to Jere fleck (1970), the future king was taught by an expert, that is, the old Norse þulr. as keeper of traditions, he may have recited the genealogy during royal inaugurations, of which one important element was the enumeration of forefathers. Most likely, he fulfilled a function also in other cultic contexts, for instance, when celebrating ceremonial feasts in the hall (sundqvist 2003a, 2003b, 2007; see however also Tsitsiklis’s (2017) very critical and thorough investigation).
in Hávamál st. 111 we find the expression ‘á þular stóli’ (on the chair of the þulr). it is interesting to note that the speaker (i.e., the þulr) here is performing his speech in Háva hǫll ‘the High one’s hall’ (probably Valhǫll, since Urðar brunnr is located close to it). The speaker states: ‘hlýdda ec á manna mál; | of rúnar heyrða ec dœma, né um ráðom þǫgðo’ (i heard the speech of men; i heard talk of runes nor were they silent about good counsel). This and other passages in Hávamál indicate that the þulr was some kind of ceremonial leader or public orator, who was perhaps also knowledgeable about runes (see below).
The þulr seems, however, to be a multifunctional type of public person who plays many roles within society. in skaldic poetry, such as Haukr’s Íslendingadrápa st. 18, and a lausavísa (st. 29) by Rǫgnvaldr jarl, both dated to the twelfth century, the term þulr signifies ‘skald’. in other contexts, it refers to a more general sort of wise person. The wise speaker in Hávamál st. 134, for instance, is called ‘hárr þulr’ (the grey-haired þulr).20 in Vafðrúðnismál st. 9, it is the wise giant Vafðrúðnir who is called ‘inn gamli þulr’. it seems that a þulr was, as mentioned, skilled in the art of runes since in Hávamál st. 142 it is said that the mighty sage (the þulr or Óðinn) painted runes (‘fimbul þulr faði’).
in other areas of Northern Europe, words equivalent to þulr had other connotations. in one glossarium, old English þyle glosses latin orator, ‘speaker, orator’, and in another it glosses latin scurra ‘joker’, ‘jester’ or ‘practical joker’
20 ottar grønvik (1999) has argued that the ek-person in Hávamál st. 111–64 is consistently a reference to the old þulr and not to Óðinn, as has previously been presumed.
758
Olof Sundqvist
(Vogt 1927, 1942). The þyle Unferð in Beowulf is an orator, spokesman, or perhaps a principal court officer.21 The poem (1165–68) states:
swylce þær Unferþ þyle
æt fotum sæt frean scyldinga;
gehwylc hiora his ferhþe treowde,
þæt he hæfde mod micel,
þeah þe he his magum nære
arfæst æt ecga gelacum.
(There too sat Unferð the spokesman, at the feet of the scyldingas’ lord [King Hroðgar]; all of them relied on his bold spirit, believing that he had great courage, although in the play of sword blades he had shown no mercy to his kinsmen).
obviously, Unferð was a great warrior besides being a spokesman. Previously, he had killed his brother, and he lent his wondrous sword Hrunting to Beowulf for the fight against grendel’s mother.
Proto-Nordic þewaz** **
The term þewaz appears in some Proto-Nordic runic inscriptions. it is usually interpreted as ‘servant’, ‘slave’, or ‘follower’ (cf. old English þēow; gothic þius) (Düwel 1992b; Brink 2003a, 2012: 127). it sometimes appears, however, in connection with old germanic names of people of high rank. We have the inscription owlþuþewaz (DR 7, Samnordisk runtextdatabas on an expensive ferrule from Thorsbjerg, angeln, schleswig (third century)). in this case, þewaz is the second element of a compound where the first element is a form of the name Wulþu. This name (or appellation) could be compared to gothic wulþus, meaning ‘splendour, brilliance’. it can also be related to the old Norse name of the god Ullr. The compound can therefore be interpreted as ‘the servant of Ullr’ (antonsen 1975; Moltke 1985; T. andersson 1993; Brink 2003a, 2012; Nordberg 2006b; laur 2009; for a different interpretation, see è5). Krause and Jankuhn (1966) argue that the o-rune at the beginning of the inscription should be interpreted as an ideograph (Begriffsrune), signifying ōþila
‘property’. When the rune-carver who carves his name onto the sword puts the o-rune in front of his name, he indicates that the sword belongs to him.
Thus, þewaz in this inscription was probably not a conventional ‘servant’ or
‘slave’ (spurkland 2005). Perhaps he was considered a high-ranking religious specialist who served a deity (i.e., Ullr). in this context, stefan Brink (2003a) has discussed several germanic compound names, which include equivalents 21 on the þyle Unferð’s role in Beowulf, see also Enright (1996a) and the overview in Tsitsiklis (2017: 9–15, 151–260). on Beowulf in general, see gräslund 2018.
29 – Cultic Leaders and Religious Specialists
759
to the term þewaz as last element. The name of the Vandal Gotthæus (gothic
* Guþþius) and the old High german name Gotadeo both mean ‘servant of god’.
The old English name Incganþeow attested in Widsið means ‘servant of the god ing’, and old High german Irmintheo means ‘servant of the god Ermin/irmin’.
Brink argues that þewaz in pre-Christian time sometimes designated a free servant, perhaps devoted or dedicated to a specific deity, and who probably enjoyed high social rank. some of the names have first elements indicating a relationship to the divine world, such as old High german Ansedeus, which includes a word related to old Norse áss ‘god’. Whether þewaz could be connected to religious functions is somewhat debated, however (see, e.g., T. andersson 1993; Wulf 1994). There are also dithematic names, including þewaz, which lack a first element referring to a religious sphere. an argument against the interpretation ‘the servant of Ullr’ in the Thorsbjerg inscription is that we should expect a genitive in the first element of the compound (see Brink 2012: 155).
Cultic Leaders ( goðar /*gudhar / goþar ) in Viking Age Sources ( c. 800–1100 ce)
The goðar** of Norway and Iceland **
Early scandinavian chieftains were sometimes called goðar (singular goði). and the authority or office they held was called a goðorð (literally ‘god’ + ‘dignity’; cf. mannaforráð ‘power, rule over people’).22 a holder of such an office was called goðorðsmaðr or goði. The term goði is derived from old Norse goð ‘god’, thus indicating an original cultic function of these leaders (cf. gothic gudja).
Besides his religious tasks, the goði also performed other societal functions, such as lawman, but also as a more general political leader: that is, he was a chieftain.23 Thus, he should be regarded as a cultic leader.
Medieval prose texts report that goðar in Norway had a close relation to the pre-Christian sanctuaries. Landnámabók (s 297, H 258) states that:
‘Þórhaddr enn gamli var hofgoði í Þrándheimi á Mæri. Hann fýstisk til Íslands ok tók áðr ofan hofit ok hafði með sér hofsmoldina ok súlurnar; en hann kom í stǫðvarfjǫrð ok lagði Mærinahelgi á allan fjǫrðinn ok lét øngu tortíma þar nema kvikfé heimilu.’
22 on goðar in general, see mainly Ebel (1998), Jón Viðar sigurðsson (1999), Dillmann (2006: 312–17). for the religious aspects of goðar, see sundqvist (2003a, 2003b, 2007).
23 it is an old scholarly tradition to refer to the goðar by means of the English term ‘chieftains’, although the semantic correspondence may, in fact, not be all that exact. see Byock (2001: 13)
760
Olof Sundqvist
(Þórhaddr the old was a hofgoði in Trondheim in Mære. He wanted to travel to iceland, and before he went, he took down the hof and brought with him the soil from the sanc tuary and the high-seat posts. He came to stǫðvarfjǫrðr and proclaimed the Mære-Peace over the entire fjord area, and nothing was allowed to be killed there, except the animals on the farm.)
This short passage gives us some important information. it tells us that Þórhaddr controlled and even regarded the cultic building ( hof) of Mære as his own property. it seems that he could do whatever he wanted with it. When he moved to iceland, for instance, he took down the building and brought the most essential parts of it with him to the new land. When he came to iceland, he also proclaimed the Mære-Peace ( Mærinahelgi) over his new land area and prescribed the ritual regulations which were to be adhered to there. The close relation between him and his sanctuary is indicated by his title: hofgoði ‘sanctuary chieftain’. Most likely, Þórhaddr was regarded as a political leader or chieftain as well. The skarðsárbók and Þórðarbók versions of Landnámabók report that he was described as a (great) chieftain ( hǫfðingi (mikill)) ( Landnámabók p. 307 n. 12).
archaeology may partly support the information in Landnámabók. Under the medieval church of Mære, remains of a Viking age building were found, which has been interpreted as a hof-sanctuary (lidén 1969, 1999). inside the building, gold foil figures were discovered. They constitute strong indication of cultic actions as well as the presence of a political-religious ruling power at this site during the late iron age. Perhaps the gold foils should be related to a wealthy family who had great ambitions of gaining power in inner Trøndelag.
since Mære was probably no magnate farm during the early Viking age, it has been suggested that the family who controlled the sanctuary lived at Egge.24 at this place, rich late iron age burials have been located.
Landnámabók combined with the archaeological evidence inform us that goðar and local chieftains played an important role in the public cult in Norway during the late iron age. There was, however, a radical change to this situation when the central royal power started to exert influence on local chieftains in the late ninth century.25 sources report that many goðar and chieftains emigrated 24 see the discussions in, e.g., stenvik (1996); Røskaft (1997: 237; 2003: 138–39); lidén (1999: 45).
25 it is possible that hofgoðar were important also in late Viking age Norway. one manuscript of Heimskringla mentions that during guðbrandr’s life ( c. 1000 ce) there was a hofgoði of the Dalesmen, who was called Þórðr ístrmagi. in the other manuscripts, Þórðr is called hǫfðingi.
see Phillpotts (1912–13: 271).
29 – Cultic Leaders and Religious Specialists
761
to iceland with their sanctuaries, in a fashion similar to Þórhaddr inn gamli.26
The reason for doing this was, according to these narratives, the harsh reign of King Haraldr hárfagri. one example is the chieftain Hrólfr Mostrarskegg, who brought his hof-sanctuary from Mostr island in south-western Norway to western iceland after a conflict with King Haraldr (see below).
There is little evidence of a formal religious organization or priesthood which owned lands in early Viking age Norway or other parts of scandinavia.
in the central settlement districts of Norway, land was owned by private persons or families. if a chieftain erected a ceremonial building on his land, it was his own property, and he could do whatever he wanted with it. These sanctuaries were thus associated with aristocratic centres and chieftains’ farms.
Þorbjǫrn hersir in fjalafylki, for instance, tended (infinitive varðveita) the hof-
sanctuary at his farm in gaular, while guðbrandr hersir cared for his hof in guðbrandsdalir at the end of the tenth century ( Njáls saga ch. 87–88; Óláfs saga helga ch. 112; Flateyjarbók, ii, 189). The people living in the settlements could, however, be invited to take part in ceremonial feasts at these sanctuaries.
Perhaps they paid tribute ( hoftollr) for this privilege. it is therefore possible that public feasts occasionally took place in private sanctuaries. Most likely, there were also some kind of communal cultic sites in Norway and sweden, which may have been organized by a cooperation of chieftains. Mære was probably a sanctuary of this type during the tenth century (that is, after Þórhaddr left Norway) (see sundqvist 2016: 172–74, 510–13).
several old Norse prose texts report that many goðar left Norway and migrated to iceland during King Haraldr’s reign. Upon arriving in iceland, they consecrated the land where they intended to build their farms (see sundqvist 2016: 291–95). it seems as if the intention was to establish some sort of sacred or ritual landscape. on their new farms, they also erected hof-buildings wherein they placed ritual objects, such as the high-seat posts. The close relationship between goðar and their hof-sanctuaries in iceland is attested in Úlfljótslǫg (Jón Hnefill aðalsteinsson 1998: 35–56), which has been preserved in the Hauksbók redaction of Landnámabók (H 268), Þáttr Þorsteins uxafóts (fourteenth century) ( Flateyjarbók i, 249), and the codex Vatnshyrna’s version of Þórðar saga hreðu 1 (thirteenth to fourteenth centuries). in what follows, the passage from 26 see, e.g., Landnámabók sH7–8; H11; s289 H 250; s310 H270. so did, e.g., Hrafn enn heimski and his son Jǫrundr goði in s 338 H 296; Kolgrímr hinn gamli in H 22; the brothers Eyvindr vápni and Refr enn rauði in s 267 H 229; Bárðr blǫnduhorn in s 340 H 298. Phillpotts (1912–13) noted that many goðar in iceland actually came from old hersir-families in Norway. for more complete documentation, see strömbäck (1928b) and Birkeli (1932).
762
Olof Sundqvist
Landnámabók is paraphrased. The texts narrates that a gold ring weighing two ounces or more should lie on the stalli (altar or platform; see below) of every chief hof-building, and, moreover, every chieftain ( goði) should carry the ring upon his arm during all public law-assemblies at which he was head of affairs, having first reddened it in the blood of an ox which he himself had sacrificed there. The text further says that the land (i.e., iceland) was then divided into quarters, and it was decided that there should be three assemblies in each quarter and three hof-buildings in each assembly district or community. Moreover, men should be selected according to wisdom and righteousness to have ward of the hof-buildings, and these men were to nominate courts of judges at the assembly as well as regulate the proceedings of lawsuits, and therefore were they called goðar. finally, every man should pay toll to the hof-building as they now pay tithes to the church.
The historical source value of this text has been much debated. olaf olsen (1966: 34–49), for instance, questioned several details in this description, such as the organization and the number of the thing-assemblies and hof-sanctuaries in iceland as well as the information about sanctuary dues ( hoftollr) and the oath-rings. according to him, the information that the goðar were ‘selected according to wisdom and righteousness to have ward of the hof-building’ must be considered nothing more than a myth, which had nothing to do with historical reality. The goðar were never selected by the people as ‘temple superintendents’. it was individuals who came from the foremost and noblest families in iceland who had the power position and possibilities to play a central role at the judicial courts and in public cult. Usually, they erected the hof-buildings on their own farms.
The Inheritance of Cultic Leadership
olsen’s argument is well founded. The free people of iceland did not select somebody to the office of goðorð or appoint a candidate to the ward of the hof-
building just because of their wisdom and righteousness, as stated in Úlfljótslǫg.
These qualifications were of course expected from a legitimate leader alongside other qualities, such as generosity and ability to obtain support from friends and allies (Jón Viðar sigurðsson 1999). There were probably also other aspects which qualified someone to the status of goði. according to some traditions, it seems that the role of cultic leader in iceland, including the charge over the hof-
building, was inherited within the chieftain family. These traditions intimate that the dignity of cultic leader was sometimes brought from Norway to iceland.
Landnámabók (s85, H73) and Eyrbyggja saga ch. 3–4, for instance, report that
29 – Cultic Leaders and Religious Specialists
763
the chieftain Þórólfr Mostrarskegg maintained the office of cultic leader after he left Mostr in Norway and settled at Breiðafjǫrðr in western iceland.
in Landnámabók, we do not get much information about Þórólfr’s family or what kind of office he held. it says only that he was the son of Ǫrnólfr fiskreki and that he lived in Mostr. it is also mentioned that: ‘hann var blótmaðr mikill ok trúði á Þór’ (he was much devoted to offering up sacrifices, and believed in Þórr). according to Eyrbyggja saga ch. 3, Þórólfr was a chieftain of considerable standing (‘hǫfðingi mikill’). it also says that he was a close friend of Þórr (‘mikill vinr Þórs’) and watched over a Þórr’s hof (‘varðveitti […] Þórshof ’) at his farm in Mostr island. No text explicitly states that he occupied the office of goði, even if he was a religious man. Evidence in both Eyrbyggja saga and Landnámabók indicates, however, that Þórólfr held some kind of office, which also included the charge of the hof-building. This office was inherited within his family by his sons and grandchildren.27 Both Landnámabók and Eyrbyggja saga mention that several male members of his family were titled goðar, that is, they were regarded as cultic leaders. in Eyrbyggja saga ch. 3, for instance, it is said that Þórólfr had a son called Hallsteinn, who was born in Norway.
in Chapters 5 and 6, it states that Hallsteinn journeyed to iceland together with Bjǫrn Ketilsson, and that he considered it a slur on his manhood that he should have land granted to him by his own father, Þórólfr at Hofstaðir, so he crossed over to the other side of Breiðafjǫrðr, to a place called Hallsteinsnes at Þorskarfjǫrðr, and staked his claim there. later in the text, Hallsteinn is called
‘goði af Hallsteinsnesi’ ( Eyrbyggja saga ch. 48). similar information is also provided in Landnámabók where he is called Þorskarfjarðargoði (M 25, s85; cf. Íslendingabók ch. 4). in this text, it is said that he sacrificed to Þórr because he wished that the god would send him his high-seat posts. after some days, a big tree came ashore, which Hallsteinn used for high-seat posts (s123, H95).
Most likely, Hallsteinn was qualified to hold a goðorð since he was a son of the chieftain and cultic leader Þórólfr.
Even if it is not explicitly mentioned, Eyrbyggja saga intimates that also Þórólfr’s other son, Þorsteinn þorskabítr Þórólfsson, was a kind of goði and took care of the hof-building at Hofstaðir after his father’s death. This can be deduced because the context of the narration suggests that Þorsteinn lost full custody over this sanctuary. according to Chapter 9, Þorsteinn had a great conflict with Þorgrímr Kjallaksson, and this conflict was resolved in such way that Þorgrímr 27 several scholars have regarded Þórólfr as a goði; see, e.g., Wessén (1924: 170); Baetke (1942a: 133–34); strömbäck (1975: 41–42). Cf. DuBois (1999: 65–66); sundqvist (2007: 25–28). Cf. sundqvist (2016: 176–80).
764
Olof Sundqvist
was to bear half of the cost of maintaining the hof-sanctuary at Hofstaðir and also that he and Þorsteinn were to share the sanctuary dues and the support of the farmers equally between them. Þorgrímr was, moreover, supposed to back Þorsteinn in all his lawsuits and safeguard the sanctity. He was from then on designated Þorgrímr goði. The text clearly intimates that, before the conflict with Þorgrímr, Þorsteinn was the only person who occupied the cultic office at Hofstaðir. Probably, he inherited this office from his father Þórólfr.
The interpretation that Þorsteinn acquired chieftaincy ( goðorð) and the charge of the hof-building at Hofstaðir by means of the inheritance from his father harmonizes with the continuation of the saga, because it is said that Þorsteinn later built a great farm Helgafell where he erected a hof-building.
at his new farm, Þorsteinn and his wife had a son called grímr. it seems that grímr inherited his father’s cultic role and the charge of the sanctuary: ‘Þann svein gaf Þorsteinn Þór ok kvað vera skyldu hofgoða ok kallar hann Þorgrím’
(Þorsteinn dedicated this boy to Þórr, calling him Þorgrímr, and said he should become a hofgoði) ( Eyrbyggja saga ch. 11; è31).
Eyrbyggja saga ch. 15 narrates further that Þorgrímr Þorsteinsson, in turn, had a son called snorri. He inherited his father’s farm and the sanctuary at Helgafell.
He was now in charge of the hof-building and was thus called goði: ‘Hann varðveitti þá hof; var hann þá kallaðr snorri goði’. We may note that the scribe of the saga uses the verb varðveitta ‘be in charge of, take care of ’ when expressing the relation between the cultic leader and his sanctuary. This expression is also used when describing Þórólfr’s relation to his sanctuary in Mostr; that is, Þórólfr watched over the Þórr’s hof (‘varðveitti […] Þórshof ’) located there.
The unknown author of Eyrbyggja saga intimates that various members of Þórólfr’s family for several generations occupied a religious office, which included the charge of hof-buildings in the areas around Breiðafjǫrðr. it also appears that the sons and grandsons of Þórólfr were devoted to the cult of Þórr.
some of Þórólfr’s sons carried, like their father and grandfather, the name of the god as a first element in their names, for example, Þorsteinn and Þorgrímr.
Whether this naming custom within goði-families really reflects ancient conditions is much debated (see, e.g., andersson 1992b; Vikstrand 2009a).
The idea of presenting Þórólfr and his descendants as a kind of ‘priest-family’ who cared for local sanctuaries at Breiðafjǫrðr was not a literary construction by the author of Eyrbyggja saga. Most likely, he based this idea on an older tradition. also in Landnámabók we see that several descendants of Þórólfr were titled goði. Þórólfr’s son Hallsteinn was called Þorskafjarðingargoði, while his grandson was called Þorgrímr goði Þorsteinsson þorskabíts and his great grandson snorri goði Þorgrímsson (s86, H74, M25).
29 – Cultic Leaders and Religious Specialists
765
The sources report that the goðorð could also be acquired in other ways as well. it could be shared between two persons, received as a gift, or even pur-chased. in, for instance, Hrafnkels saga Freysgoða ch. 4, it is mentioned that Þorkell leppr Þjóstarsson handed over his position of authority ( mannaforráð) to his brother Þorgeirr when he went abroad. Before that, he had been a goðorðsmaðr. The text intimates that he originally inherited the title goði from his father. in the course of events, Þorgeirr offers to return the goðorð and the position of authority to his brother Þorkell for a period, but then he suggests that they could share it equally between them. Þorkell, however, refuses to take this offer from his brother, since he feels that Þorgeirr is the more accomplished of them both for this task. also in Eyrbyggja saga it is said that Þorgrímr and Þorsteinn shared the goðorð at Þórsnes. We must not forget that these statements appear in medieval texts. Most historians accept, however, that the goðorð could be handled in such a way (see, e.g., Byock 2001).
The sources indicate that the icelandic goðar should not be seen as persons who exclusively occupied a priestly office and permanently appeared in cultic roles (cf. gunnell 2001c). Most of them functioned as general political-judicial leaders as well and also had many secular leadership functions within society.
They could thus be described as both hofgoði and lagamaðr. some of them were, however, closely associated with a specific deity. it could, for instance, be Þórr,28
but it could also be freyr.29 in such cases, the chieftains’ sanctuaries were sometimes dedicated to the same deity.
Ritual Objects and Symbols of Dignity
it seems as if goðar also carried certain ritual objects and symbols of dignity.30
as was mentioned above, Úlfljótslǫg states that every man who was to transact any business at an assembly must first swear an oath upon a certain ring, name two or more witnesses in evidence, and then speak the following words: ‘at ek 28 other than Þórólfr and his family, we may refer to Þorgrímr goði, who had a hof-sanctuary at Kjalarnes in southwestern iceland in which Þórr was the most worshipped god ( Kjalnesinga saga ch. 2).
29 see, e.g., the traditions about the chieftain Hrafnkell freysgoði in Hrafnkels saga Freysgoða ch. 2. Landnámabók mentions a chieftain called Þórðr freysgoði Ǫzurarson; see H 276.
30 Even if we have no clear evidence for an initiation ritual on the goðar in the sources, it seems natural from a comparative perspective that these symbols of dignity were received at such a ceremony after a period of training into the cultic office. in the mythic accounts we may see traces of initiations (see schjødt 2008; sundqvist 2009b, 2010).
766
Olof Sundqvist
figure 29.3. a panel on the picture stone from Tängelgårda on gotland (sHM 4373:108186) depicts a procession with men holding rings in their raised hands. The picture stone is dated to the ninth or tenth century. Photo: statens Historiska Museum, stockholm.
vinn eið at baugi, lǫgeið; hjálpi mér svá freyr ok Njǫrðr ok hinn almáttki áss’
( Landnámabók H 268) (i take an oath upon the ring, a lawful one, so help me freyr and Njǫrðr and the all-powerful god). a similar description of oath-rings and hof-buildings is also found in Eyrbyggja saga ch. 4. Here, it also says that a structure of some sort (old Norse afhús) was built inside the hof of Þórólfr’s farm at Hofstaðir. inside this recess, there was an altar or platform in the middle, called stalli. on this platform lay an open ring without joints ( hringr einn mótlauss), weighing twenty ounces, upon which people had to swear all their oaths. it was also the duty of the hofgoði to wear this ring on his arm or in his hand at every public meeting. similar information appears in Kjalnesinga saga ch. 2, where it is mentioned that a ring made of silver was placed on the platform in Þorgrímr goði’s hof-sanctuary at Kjalarnes. When an assembly was held, the hofgoði had this ring in his hand. People also swore oaths on this ring.
in these texts, the rings are described as ritual objects belonging to the sphere of goðar. in connection with them, rituals were performed which seem to link to the sacrifice, but also to judicial matters, such as oaths. Most likely, these rings were also symbols of honour linked to the goðar. When appearing in public
29 – Cultic Leaders and Religious Specialists
767
functions at sacrifices or judicial assemblies, the goði must carry the ring in his hand or on his arm. These descriptions in the sagas of icelanders have, however, been debated. olaf olsen (1966) argues that they do not say anything at all about pre-Christian religion and ancient customs in iceland. Nor was Úlfljótslǫg, according to olsen, based on ancient tradition. The oath-formula, for instance, with the expression ‘hinn almáttki áss’, which he interprets as ‘the almighty god’, instead had a Christian origin (cf. Kabell 1975 and è31; see above).
But perhaps olsen’s and other scholars’ source criticism has been too severe. There is, in fact, information in some more trustworthy sources which appears to confirm at least part of the information from the sagas. There are, for instance, some passages in eddic poetry where the custom of swearing oaths on rings is validated. in Hávamál st. 110, a baugeiðr ‘ring-oath’ is mentioned, and in Atlakviða in grœnlenzka st. 30 it is stated that oaths were sworn on Ullr’s ring (‘eiða opt um svarða […] oc at hringi Ullar’). There is also a note in the anglo-saxon Chronicle dated to 876 which is of great interest in this context. it states that western scandinavian Vikings swore oaths on a sacred ring and promised to refrain from ravaging within the realm of King alfred (quoted in Kabell 1975: 36).
Today, we have archaeological evidence of ‘oath-rings’, such as the big ring ( c. 43 cm in diameter) from forsa in Hälsingland, sweden ( figure è20). it is decorated with runes and has for a long time been in the possession of the church in forsa. When it first came to the attention of scholars in the eighteenth century, it was nailed onto the door between the porch and the nave inside the old church, which was pulled down in 1840. for a long time, there was a consensus among scholars that the runic inscription on this ring was a medieval clerical legal enactment from the twelfth century (è20). Recently, however, it has been argued that the ring should not be regarded as a clerical record (liestøl 1979; Ruthström 1990; Brink 1996b, 2010b; Widmark 1999; Källström 2007, 2010b, 2011; sundqvist 2007, 2016, 2017; Williams 2008), because the content of the inscription is, instead, a Viking age enactment, more precisely from the ninth or tenth century. also the function of the ring has been reinterpreted. in previous research, it was thought that the ring should be seen as a ‘church-door ring’, whereas today, scholars argue that the forsa-ring was an ‘oath-ring’. This ring was probably kept at the important assembly place (and cultic site) of Hög. The persons mentioned in the runic inscription, anund in Tåsta and ofeg in Hjortsta, may very well have been a sort of old swedish * gudhar (cf. old Norse goðar) with both religious and judicial functions within the local area. Maybe they carried the ‘oath-ring’ in their hands during religious ceremonies, as a sign of dignity and religious authority.
768
Olof Sundqvist
a picture-stone from Tängelgårda, gotland, may display just such cultic leaders taking part in a procession and carrying ‘oath-rings’ in their hands.
The information in Eyrbyggja saga ch. 4 that the ring must be made without joints ( mótlauss) is interesting. That some symbolic rings must be without joints is attested from elsewhere. in connection with the small cultic house in Järrestad, eastern skåne, which is located next to a multifunctional hall, a ring was discovered (söderberg 2005). This ring had a diameter of 9.5 cm. it was forged from round-iron and a break was subsequently made to give the ring an opening. This feature could be associated with the icelandic ‘oath-rings’, which must not be closed when carried by the goðar during sacrificial ceremonies. it could perhaps also be associated with the ritual restrictions which the Roman Jupiter-priest flamen dialis was required to follow: he was not allowed to carry closed rings (aulis gellius, Noctes Atticae 10.15.7).
*Gudhar/goþar** in Eastern Scandinavia **
Placenames and runic inscriptions indicate that an office equivalent to old Norse goðar, namely, old swedish *gudhar and old Danish goþar, existed in east scandinavia (e.g., Hellberg 1986a; Brink 1996a; Vikstrand 2001). it seems that they, too, were closely linked to public sanctuaries. The close connection between the *gudhi- office and public sanctuaries appears to be evidenced in the placenames around lake Mälaren. The farm name Gudby (< Gudhaby), in fresta parish, Uppland, for instance, has been interpreted as ‘the *gudhi’s farm’.
The placename specialist lars Hellberg (1976, 1986a) has suggested that the area around Gudby was the cultic centre for the people living in the settlement districts or ‘small regional formation’ called *Valand (Vikstrand 2001: 387–88).
Directly north of the village, there is a settlement (farm) designated Vallen sjö.
according to Hellberg, that name is derived from the name *Valændasior ‘the lake of the inhabitants of Valand’. The first element includes a genitive form of the plural inhabitant designation valændar, which is derived from the settlement name *Valand. This lake, which today has been drained, probably played a significant role for all people living in *Valand. Hellberg argues that this lake was regarded as a holy lake and that the *gudhi in the neighbouring village may have functioned as ‘hednisk präst’ (religious specialist) for all valændar when performing sacrifices at this place (cf. Brink 1997: 430).
it seems, thus, that the old swedish *gudhar sometimes were in charge of public cult for local or minor regional groups within the lake Mälaren region, such as the valændar. it is possible that East scandinavian *gudhar/ goþar had many functions within society in a way similar to the icelandic goðar. They could,
29 – Cultic Leaders and Religious Specialists
769
for instance, bear various titles simultaneously. The glavendrup inscription (DR 209, Samnordisk runtextdatabas) from fyn, Denmark (tenth century) is an example of this phenomenon. Erik Moltke (1985: 226) reads and interprets one sequence of it thus: **… auft | ala . saulua kuþa | uia l(i)þs haiþuiarþan þia | kn **
… ‘… in memory of alle, goþi of the sølver, honour-worthy thegn of the uia-host
…’. according to Moltke, alle was not only the old Danish goþi of the inhabitants called sølver, he was also an honour-worthy thegn, that is, a free man and successful ‘warrior, champion’ of high rank.31 if we accept Jan Paul strid’s (1999) interpretation of the runic inscription of the Karlevi stone (Öl 1, Samnordisk runtextdatabas), in Öland ( c. 1000), we may also there meet a multifunctional leader designated *gudhi, old Danish goþi (critically considered by Marold 2000b). in the Karlevi inscription, ‘sibbi the goþi’ is, after his death, honoured with a complete dróttkvætt stanza wherein he is praised as a great and upright sea warrior, who ruled ‘over land in Denmark’ (Jansson 1987; Düwel 2008).
The inscription at gursten (sm 144, Samnordisk runtextdatabas), småland, is also noteworthy in this context. The placename derives from old swedish Godhastæinn, that is, ‘stone of the *gudhi’ or ‘part of † Sten that belonged to the *gudhi’ (Hellberg 1979. Cf. Brink 1999; strid 1999). it is possible that a lost hamlet close to Gursten was named † Sten, a name possibly denoting the hillfort situated in the vicinity. otto von friesen (1914) dates the inscription to the late ninth century. The first part of the sequence kuþaskaki in line 3 of the inscription can be interpreted either as the personal name Gudhi in the genitive case or as a genitive form of a common old swedish noun (appellation) *gudhi (sg. or pl.) (see Kinander 1935–61: i, 292–96). This personal name is not very well documented.32 We may therefore assume that the term
*gudhi here denotes a person with religious leadership roles, that is, it is used as a title (von friesen 1933). Perhaps skeggi’s father was a *gudhi or he came from a family of *gudhar (Källström 2007). Perhaps this family previously had military functions at the hillfort ( Godhastæinn), beside their religious assignments. The picture of the east scandinavian *gudhar/goþar that can be pieced 31 Moltke argues that þegn in this context ‘denoted a kind of military status. “Thegn” is then a title of rank’ (1985: 286). Michael lerche Nielsen (1998) opposes this interpretation and states that þegn could just as well mean ‘free yeoman’ in this context, and thus this inscription would not support the notion that alli was a multifunctional leader (cf. Nielsen 1968: 10–16; lindow 1976: 106).
32 it may appear in a few inscriptions (sm 96; Vg 187; U 579, Samnordisk runtextdatabas).
ingrid sanness Johnsen states that the proper nouns Góði and Guði as well as the appellative
*gudhi/ goði are theoretically plausible (Johnsen 1968: 162).
770
Olof Sundqvist
together from runic inscriptions is very uncertain, however. We cannot tell from the inscriptions alone whether these *gudhar/goþar appeared in multifunctional leadership roles and as general chieftains. since the related old High german term gotinc/ goting appears as a gloss for tribunus (chieftain) (see above) and thus indicates a picture which harmonizes with the information on old Norse goði provided by the medieval icelandic texts, we may assume that also the East scandinavian * gudhi/ goþi was a chieftain who carried out many societal functions. such chieftains appear to be linked to local or regional sanctuaries where they also organized public cultic activities for specific groups of people, most likely the local inhabitants. Most of these sanctuaries were probably located at the chieftains’ farms, but some might have been regarded as communal cultic sites.
Religious Specialists in Viking Age Sources?
Old Norse Ǫlvir** **
according to sigvatr Þorðarson’s poem Austrfararvísur, composed around 1020, the Christian skald travelled to sweden (‘til svíþjóðar’) in the beginning of the eleventh century. finally, he arrived at Hov (‘til Hofs’) (perhaps in Dalsland), but the people there turned him away, since that day (or place; see Austrfararvísur st. 4) was to be kept holy ( heilagr). The people were celebrating álfablót, that is, sacrifices to the divine beings called álfar (è31) and (è63).
a woman told the skald in st. 5: ‘vér erum heiðin’ (we are heathen people) and that ‘ek hræðumk við reiði Óðins’ (i fear the wrath of Óðinn). it is notable that three of the farmers who drove sigvatr away are referred to as Ǫlvir (see st. 6). The second element in this name, which may also be a title, consists of old scandinavian -vé(r), -vi(r)/ -væ(r) (Proto-Nordic * wīhaz). John Kousgård sørensen (1989) argues that some composite words including this element ( *wīhaz) may indicate a pagan priestly office. sometimes, these composites have guð ‘god’ as first element, for example, Runic swedish Guðvēr. in other cases, the first element is a name of a deity, as for instance in swedish Tore (cf. old Norse Þóri[r]). The first element occasionally consists in a designation for a cultic place, such as Al-, Sal-, Vi-, Hargh-, as for instance the name or title Salvir, which has been interpreted as ‘priest of the sal-sanctuary’. The name Ǫlvir belongs to this group. Jan de Vries (1932; 1956–57a: i, 402; 1962a: 687) interprets it as *alu-wīhaz ‘priest of an alu-( alh-) sanctuary’ (Priester eines alu-( alh-) Heiligtums). it should be noted that the first element *al, *alu, is much debated (Brink 1992a; Vikstrand 2001; Elmevik 2004). However, the composites including this ele-
29 – Cultic Leaders and Religious Specialists
771
ment ( *wīhaz) undoubtedly refer to a person of religious significance. Whether the three named Ǫlvir in Austrfararvísur should be regarded as cultic leaders or religious specialists must, however, remain uncertain.
Old Swedish *lytir**/Old Norse Lytir**
lennart Elmevik (2003b; cf. 1966, 1990) has suggested that old swedish
*lytir, which appears as first element in some swedish placenames — for example, Lytisberg in Östra Husby, Östergötland, Lytislunda in Österrekarne, södermanland, and Litslunda, in lillhärad, Västmanland — originally derived from a Proto-Nordic *hluti-wīhaz ‘diviner, truth-teller, sacrificial priest’.33 in some cases, the first element may just as likely be derived from a word related to old Norse hlutr (old swedish luter, loter) ‘lot which is used in divinations’. The appellation *lytir may thus be interpreted as a religious specialist particularly associated with divination rituals (cf. Nygaard and Murphy 2017: 67–68). old English hlȳtere (cf. tān-hlȳtere) — which is a gloss for clericus, meaning ‘clergy-man, priest’ — supports such an interpretation (Clark Hall 1916). according to Rimbert’s Vita Anskarii ch. 18, there was a diviner (latin divinus) who performed divination rituals among the svear in Birka c. 830–40 ce. Perhaps the native term for him was *lytir, that is, an expert in casting and interpreting lot-staves ( fella blótspán; hlauttein hrista), observing the flight or sound of birds, and so forth. one reservation concerning *lytir, however, is that Lytir may be the name or byname of a god. The name is used specifically with reference to a deity in Hauks þáttr hábrókar in Flateyjarbók (strömbäck 1928a).
Thus a possible interpretation is that Lytisberg, Lytislunda, and so forth should be regarded as theophoric place-names, with the name of this god as the first element (Vikstrand 2001).
Old Swedish *vivil**, Runic Swedish Vivil****, Old Norse Vífill**** **
Equivalent forms to the old Norse name Vífill are attested in swedish runic inscriptions and place-names (Runic swedish/old swedish Vifill/Vivil/* vivil) 33 simon Karlin Björk (2015) has recently discussed the sequence lutaris, representing a personal name in the genitive form, which is found on the Danish rune stone DR 145. He interprets the name as an old Danish * Lutari( r), from the word hlutr ‘lot, share’, or alternatively * Lȳtari( r), from the verb hljóta ‘get as one’s lot’. He argues that this name was originally a common noun describing a person with a cultic function and meaning ‘caster of lots, soothsayer’, related to old High german liozari and old English tān-hlȳtere ‘soothsayer’.
772
Olof Sundqvist
(andersson 1992b: Vikstrand 2001). it appears, for instance, in a sequence of the gotlandic runic inscription at Pilsgårds, Boge parish (g 280, Samnordisk runtextdatabas), uifil, employed either as a name or an appellative (gustavson 2001; Thorgunn snædal 2002). The farm name Vivelsta, Markim parish, Uppland, is also interesting. lars Hellberg (1986a) argued that this name derived from the form *Vivils-Husar. in his opinion, the placename designates
‘en bebyggelsepart i centralorten *Husar (* Husa), som tilldelats dess präst, kallad vivil(l)’(Hellberg 1986a: 63) (a part of the settlement in the central region *Husar, that was allocated to its priest, called vivil[l]). independently of Hellberg, gunter Müller (1968) also interprets the name/appellative Vivil/ *vivil as a ‘consecration name’ (Weihename) or the ‘title of a priest’.
Müller connected it with the Proto-Nordic diminutive * wīgwilaz ‘the little consecrated one’ (see above).34
Vivelsta in Markim parish is located in an area that is interesting from the historian of religions’ point of view. The religious character of the place may be supported by ritual depositions discovered by the lake at Vivelsta (Zachrisson 1989). The central place Husby, not far from Vivelsta, was most likely a thing-
place during the Viking Period, that is, the assembly site for the people in the area. There are several Viking age runic stones, some perhaps of a symbolic significance, appearing in Markim and orkesta in the vicinity of Vivelsta (Zachrisson 1998). it has been argued that one and the same family may have had political power and controlled the public cult during the Conversion Period, since Vivelsta is located close to the early medieval vicarage and Husby (Vikstrand 2001).
Placenames in gotland indicate that the office of *vivil also existed there.
There is a place called Vivlings, south-east of the church in Hellvi parish. The parish-name Helgawi appears in documents dating back to the fourteenth century, and it includes the words gutnish hailigr ‘holy’ and vi ‘cultic place’.
ingemar olsson (1996) carefully suggests that the placename Vivlings includes the word *vivil meaning ‘pagan priest’. The locality of Vivlings close to the old assembly place (probably Helgawi) indicates that a *vivil was situated there during the Viking age (cf. Blomkvist 2002). it is thus quite possible that the
*vivil- institution was spread across scandinavia during pre-Christian times and that the *vivil was some kind of religious specialist.
34 it should be noted that another etymology exists, which is derived from Proto-germanic *webilaz, old High german wibil, old English wifel, swedish vivel, ‘some sort of a bee-tle’ (Peterson 2007).
29 – Cultic Leaders and Religious Specialists
773
Female Cultic Leaders and Religious Specialists in the Viking Age Sources The gyðja
The sagas of icelanders also report that females, designated gyðjur (singular gyðja) or hofgyðjur, could be in charge of hof-buildings in iceland. Vápnfirðinga saga ch. 5, for instance, describes the hofgyðja called steinvǫr. she was in charge of a major hof-building (‘varðveitti hǫfuðhofit’) at the farm called Hof in Vápnfjǫrðr, in eastern iceland. all the local farmers must pay sanctuary tributes ( hoftollr) to her. We cannot rule out that steinvǫr also had a political position in Vápnfjǫrðr since she took care of the sanctuary tributes. in order to carry out this task, she was supported by the chieftain Brodd-Helgi, which indicates that she herself did not control any military or physical power.
other gyðjur and hofgyðjur also occur in the medieval texts. in Landnámabók (s180, H147) and Vatnsdæla saga ch. 35, for instance, Þuriðr gyðja sǫlmundardottir is mentioned. she was connected to the farm Hof in Vatnsdalr, where a hof-building had been erected by ingimundr inn gamli.
Þorlaug gyðja Hrólfsdóttir was, according to Landnámabók, associated with the hof-sanctuary at Reykjardalr in south-western iceland (s 41, H 29), while Þuriðr hofgyðja Véþórmsdóttir and her brother Þórðr freysgoði Ǫzurarson were linked to sanctuaries situated in Bakkárholt (H 276). Magnus olsen (1926) argues that the gyðjur were exclusively associated with the fertility cult directed towards freyr and that these female cultic leaders were regarded as this deity’s wives (Norwegian ektefelle). The weak evidence can neither deny nor corroborate this. But perhaps we can connect at least Þuriðr hofgyðja to freyr, since she was the sister of Þórðr freysgoði, whose family was called Freysgyðlingar ‘priestlings of freyr’. in the region of their home there is, moreover, a place called Freysnes (freyr’s headland). in Ǫgmundar þáttr dytts, there is a female cultic leader, perhaps a gyðja, who is called ‘freys kona’
(freyr’s wife).
a lausavísa by the skald Þorvaldr Koðránsson (tenth century) provides early and more reliable evidence of a gyðja called friðgerðr, whom we encounter in a cultic context. according to Kristni saga ch. 2 (where the verse is preserved), Þorvaldr arrived at her farm in Hvammr in western iceland together with a bishop. The two men preached the faith, while friðgerðr was inside the hof-
sanctuary, performing a sacrifice (‘enn friðgerðr var meðan í hofinu ok blótaði’).
friðgerðr’s son skeggi laughed at the missionaries. The lausavísa itself gives a vivid eyewitness description of the gyðja friðgerðr and her son skeggi. They seem to be upset, since the skald and the bishop are disturbing them with their
774
Olof Sundqvist
mission. skeggi is called ‘hlautteins hreytir’ (the one who cast lots) and ‘goða sveinn’ (servant of the gods). These expressions may not have been chosen by mere accident. Perhaps they refer to the activities that the skald really observed during his stay at Hvammr. The stanza also says that friðgerðr was shouting from the pagan ‘altar/platform’ (‘gall of heiðnum stalla’). Even if the ritual context is not fully clear, it seems that friðgerðr was performing sacrifices at the
‘stallr’/’stalli’ (altar/platform), while skeggi was casting lots (for a thorough examination, see Düwel 1985). The term stallr/stalli indicates that they were at a sanctuary, perhaps inside a hof-building. The prose in Kristni saga supports this interpretation. in the prose of Flateyjarbók (i, 270) (where the verse also is preserved), it is mentioned that friðgerðr was sacrificing during Þorvaldr’s visitation. The sources thus suggest that the gyðja and her son performed both sacrifices and divination rituals inside a ceremonial building, while Þorvaldr preached about Christ for the people in Hvammr.
The vǫlva
The female religious specialists (and mythical beings) called vǫlur (also called seiðkonur, vísindakonur, or spákonur) are well attested in eddic poetry and the sagas of icelanders, whereas they only appear once or twice in the oldest skaldic poetry35 and never in placenames. in the sagas, vǫlur often appear as ‘diviners’.
They obtain knowledge by performing a ritual known as seiðr at a ritual platform or construction called seiðhjallr (see sundqvist 2012b). Seiðr was a form of magic or divination ritual, which has also been associated with shamanism (cf. strömbäck 1935; Buchholz 1968; Price 2002; critically considered by e.g.
Dillmann 2006). The ritual practice of seiðr furthermore appears in mythical contexts, particularly in relation to Óðinn, but it also has connections to the goddess freyja ( Ynglinga saga ch. 4, 7). sometimes, vǫlur appear in other rituals, too, but it seems that they never performed sacrifices. The vǫlur moved about from farm to farm and were sometimes followed by an entourage. according to late sources, they received payment for their rituals.36 The vǫlva should thus be regarded as a professional religious specialist.
The most important account of a vǫlva appears in Eiríks saga rauða ch.
4, which tells the story of Þorbjǫrg lítilvǫlva. in this passage, her ritual dress 35 There is one attestation of a vǫlva in Kormákr’s lausavísa 48 (tenth century), and one later instance in Hofgarða-Refr gestsson’s Ferðavísur 2 (eleventh century).
36 see, e.g., Norna-Gests þáttr in Flateyjarbók (i, 358). see also Price (2002) and Dillmann (2006: 367–69).
29 – Cultic Leaders and Religious Specialists
775
and attributes are described (critically considered by Dillmann 2006; schjødt 2007c; Tolley 2009a; gardeła 2016). she wore a dark blue robe adorned with stones as well as a necklace of glass beads. on her head, she wore a black lambskin hood lined with white catskin. she carried a staff in her hand which was adorned with brass and precious stones.37 around her waist, she had a belt with a leather purse in which she kept some witchcraft accessories. on her feet, she had shaggy calfskin shoes with long straps and on her hands catskin gloves, which were white and fluffy inside. she was treated by everyone with respect.
The farmer had prepared a place for her to which he led her. on her high seat was a pillow stuffed with chicken feathers (è22) and (è30).
in the medieval traditions, vǫlur and other performers of seiðr are sometimes described in a negative sense, that is, as deviants from the social norms (e.g., Raudvere 2003; Meylan 2014a; critically considered by Dillmann 2006). They perform suspicious rituals, which are sometimes described as evil. obviously, there is a Christian perspective in these sources. This tendency may even be seen in the eddic lays. The mythic seiðr-skilled vǫlva called Heiðr, in Vǫluspá st. 22, for instance, was ‘æ var hon angan illrar brúðar’ (always the favourite of wicked women).38 in Lokasenna st. 24, Óðinn is accused of having practised seiðr and beating drums as vǫlur do. it says that he appeared in the likeness of a wizard (‘vitca líki’). The seiðr-ritual he practised is described as the ‘args aðal’ (hallmark of a pervert).39 it has been argued that the art of seiðr was a gender-determined activity, intended only for females (see, e.g., strömbäck 1935; ström 1985).
Men who performed this ritual were considered to be marred by ergi ‘sexual aberration or abnormality’, ‘inordinate sexual desire’, or ‘lasciviousness’ (cf. the adjective argr) (cf. strömbäck 1935; Meulengracht sørensen 1982; Dillmann 2006). in Ynglinga saga ch. 7, for instance, snorri says this about seiðr: ‘En þessi fjǫlkynngi, er framið er, fylgir svá mikil ergi, at eigi þótti karlmǫnnum skam-laust við at fara ok var gyðjunum kennd sú íþrótt’ (But this sorcery is attended 37 Recently, leszek gardeła (2016) has made a thorough and critical assessment of all such known staffs from the Viking age North including archaeological, iconographic, and textual sources leading to an interpretation of these staffs as ‘multivalent objects’ (2016: 219), but often with magical purposes.
38 on the different text versions and interpretation of this stanza, see strömbäck (1935).
Cf. Dronke (1997).
39 strömbäck (1935: 27). Dronke (1997: 338) translates ‘oc hugða ek þat args aðal’ as:
‘that i thought an unmanly nature’. for a critical discussion on the expression args aðal, see Dillmann 2006.
776
Olof Sundqvist
by such wickedness40 that manly men considered it to shameful to practise it, and so it was taught to priestesses).
Catharina Raudvere (2003) states that narratives involving vǫlur and seiðr include certain anomalies (cf. DuBois 1999). The performers of seiðr are, on the one hand, described as important and good for society; on the other hand, they are odd, exotic and deviant when it comes to, for instance, their very old age, eating, and clothing habits, as well as ethnicity (or geographic origin).
Especially the last aspect is used to indicate deviation in the texts. The seiðr-
performing Kotkell family in Laxdœla saga ch. 35, for instance, came from the Hebrides. in general, the seiðr-performers are said to have a finnish or sámi origin or background, and often they are females such as the seiðkona (in the text called finna) in Vatnsdœla saga ch. 10. also the vǫlva Þorbjǫrg in Eiríks saga rauða ch. 4 is exotic in some sense and deviates from the ‘normal’. she comes in a special costume with certain attributes to the farm as a temporary guest, and she brings knowledge which could be considered numinous. she is fed a meal including ritual food made from various animals’ hearts. Thus, she is ‘clearly distinguished from the other women present at the farmstead, and the details of her dress underscore her alterity’ (DuBois 1999: 124). in addition, Þorbjǫrg seems to be very old, since she is said to have had nine sisters, all of them seeresses, but now only she was still alive.
These anomalies and exotic attributes relating to the seiðr-performers could be interpreted to mean that such persons only appeared in marginal contexts, far away from the aristocratic halls and the large public blót-feasts, where mainly males perform the sacrificial rituals. as soon as seiðr-performers approach the aristocratic milieu, they seem to be portrayed as evil, at least in some sagas (see, e.g., Meylan 2014a: 49–91). one example is Hulðr in Ynglinga saga ch. 13–14, who creates a lot of trouble and much evil for the ynglinga family when she comes into contact with the royal milieu. With her seiðr-rituals she kills King Vanlandi and puts a curse on the whole family, which subsequently causes the death of King Vísburr.
The vǫlur rarely appear in the oldest written sources, such as skaldic poems or runic inscriptions. This may suggest that they were not part of the elite activities. Neither is the concept vǫlva attested in placenames, which may be taken to indicate that vǫlur never had any officially recognized position within society and that their activities during the Viking age belonged to the social periphery or in the countryside far from the central settlements. But it is also possible that the clerical scribes, who fixed the traditions on parchment, 40 The word ergi should rather be translated to ‘perversion’ here, and not wickedness.
29 – Cultic Leaders and Religious Specialists
777
deliberately described the vǫlur negatively in order to belittle them and their activities, or to turn them into something exotic. The negative view on seiðr-
performers may also be due to the distorted image projected onto them by previous research, which actually does not harmonize with the evidence from the sources. Moreover, archaeology from recent years seems to contradict the idea of vǫlur as persons at the bottom of the social hierarchy, since some women who may very well be identified as vǫlur are found in graves that are in no way poorly equipped (see, for instance, Price 2012; gardeła 2016).
françois-Xavier Dillmann has shown in his thorough investigation on magicians in iceland, Les magiciens dans l’Islande ancienne (2006), that the image of vǫlur must be reconsidered compared to the one often found in previous handbooks and specialist literature. Dillmann focuses on the texts Íslendingabók, Landnámabók, the Íslendingasögur, and the Íslendingaþættir, and states that the magicians there generally represent the ‘norm’ and that they are often well integrated and respected within society. By means of a careful reading of these texts, Dillmann is able to show that equally many men and women perform magic in these sources and that their ages vary significantly. They belong to the social group called ‘the free’, they own land, and they are frequently attached to the elite of the society. in terms of ethnicity, they usually derive from Norwegian families and have their geographic homeland within Norway. in iceland, they live within the ‘usual’ settlements and neither in caves nor on isolated islands.
Neither are they described as ‘perverted’ or ‘sexually deviated’.
The image of seiðr-performers and magicians produced by Dillmann actually harmonizes well with the picture we gain from other source categories.
archaeological finds indicate that vǫlur also existed in aristocratic milieus in both sweden and Norway, even at the very highest social stratum of society (Price 2002; gardeła 2016). in one rich female Viking age grave from Klinta, Köping parish, Öland, a sturdy staff or wand has been discovered. it was made out of iron, had a little miniature building on the top, was decorated with a little animal head, and also had a ‘basket’ close to the top. Neil Price (2002; cf. gardeła 2016) interprets it as a possible staff of a vǫlva, a vǫlr (cf. old Norse stafr, seiðstafr, gandr, gambanteinn; gardeła 2016: 135–70 critically treats the textual occurances of these terms) (cf. Tolley 2009a).41 in three chamber graves from Birka (Bj 660, Bj 834, Bj 845), staves resembling this have been found.
They have been interpreted in a similar way (Price 2002, cf. gardeła 2016).
furthermore, they appear in find contexts indicating that the users belonged 41 Eldar Heide (2006a, 2006b) suggests that these staffs should be regarded as distaffs, but with a symbolic meaning.
778
Olof Sundqvist
figure 29.4. The top of an iron
staff, with a representation of
a house. The staff was found
in a woman’s grave at Klinta in
Köpingsvik on Öland, and dated
to the first half of the tenth
century (sHM 25840:107776)
Photo: gunnel Jansson, statens
Historiska Museum, stockholm.
to the highest social stratum. These staffs are of a similar size as the one from Klinta. They all feature a knob-like mount on the shaft. a fragmentary staff was found in grave 4 at fyrkat, Denmark, and this, too, was quite a rich woman’s grave. The original length of the staff is unknown (Price 2002). The Birka staffs and the staff from Klinta were all decorated and had knobs on their shafts. This recalls Þorbjǫrg’s staff in Eiríks saga rauða ch. 4, which was ‘ornamented with brass and decorated with gemstones just below the knob’.
The wealthiest female burial in all scandinavia has been excavated in southern Norway (ingstad 1993; gansum 2002). at oseberg in slagendalen, southwest of oslo, two women were laid to rest inside a royal mound in c. 835–50.
The grave is extremely rich, and therefore it has been suggested that it was a queen with her female servant who were buried there. according to anne stine ingstad (1993), Neil Price (2002), and leszek gardeła (2016), the finds indicate that one of them was a vǫlva. They suggest that the staff found there could be interpreted as the vǫlr ‘staff ’ of a vǫlva (= ‘staff-carrier’; gardeła 2016: 154–56 is critical of this interpretation, as the staffs are most often called by other old Norse terms, as was mentioned above).42 on a carriage found in the grave, several cat symbols were carved, and in a chest cat skins were discovered.
ingstad argues that the cat and its skin were important symbols for vǫlur. she refers to Eiríks saga rauða ch. 4 where it says that Þorbjǫrg had catskin gloves on her hands; on her head she wore a black lambskin hood lined inside with 42 This interpretation is critically considered by Pesch (1999). see also arwill-Nordbladh (1998) and steinsland (2004).
29 – Cultic Leaders and Religious Specialists
779
white cat skin. in the oseberg grave, a little leather purse was found, which contained seeds from the plant cannabis sativa. according to Eiríks saga rauða ch. 4, Þorbjǫrg wore a belt made of touchwood round her middle, ‘ok var þar á skjóðupungr mikill; varðveitti hon þar í taufr þau er hon þurfti til fróðleiks at hafa’ (and on it was a big skin pouch in which she kept those charms of hers which she needed for her magic). it is thus possible that the rich woman in the grave was a vǫlva or that her female companion had such ritual functions. These finds also indicate that seiðr-performers and magicians did, in fact, appear in elite contexts.
Concluding Remarks
There may well have been some kind of ‘cultic leaders’, who were responsible for specific ritual functions at particular sanctuaries. some of them may well have taken on the role of organizers of public sacrificial feasts and as custodians of shrines and of cultic activities. These ‘cultic leaders’ (e.g., goðar) are also likely to have carried certain insignia (such as oath-rings) when performing religious duties. Whether they were permanently employed as ‘priests’ at the sanctuaries and exclusively performed religious tasks is uncertain. Most likely, many of them had functions similar to those of other leaders of society and only occasionally stepped into their cultic role. at certain important sanctuaries, such as Uppsala or lade, the ruler (i.e., the king or the earl) could have had the overarching responsibility for the public sacrificial feasts and may have played important ritual roles at these gatherings. When performing complex rituals, such political leaders probably needed assistance from other cultic leaders and/
or religious specialists (see sundqvist 2016: 189–92).
Perhaps there were also various kinds of more exclusively ‘religious specialists’ and ‘magicians’ in the Viking age, such as vǫlur and seiðr-performers. some sources indicate that they could earn their living by means of performing rituals and thus exercised religious practice as their ‘profession’. No sources support the assumption, however, that the ‘cultic leaders’ or the ‘religious specialists’
formed a closed or hierarchical organization (i.e., a priesthood) that in any way endeavoured to standardize an official world-view, a particular theology, special dogmas, or ritual practices for a wider group of people. it is likely that religious offices were commonly inherited within certain families, who predominantly cared for their own religious traditions at local cultic sites situated at these families’ farms. There are also signs indicating that cultic leaders cooperated at some communal sanctuaries.