Jens Peter schjødt
The classification of rituals was discussed above (è25), and whereas crisis rituals and cyclical rituals have not been theorized separately from a classificatory perspective, this is, on the contrary, the case with passage rituals.1 The very notion can be dated to 1909, with the appearance of arnold van gennep’s book Les rites de passage. Van gennep had a broader definition of rites de passage than the one we have chosen here (è25), since he included many rituals which we have classified as ‘cyclical’ (van gennep 1960: 178–83). Nevertheless, van gennep’s analyses of passage or transitional rituals ( rites de passage) are still important, even fundamental, to the ritual interplay between this and the other World. This is mainly due to his famous distinction between the various parts or phases of such rituals: namely, separation rites (rites symbolizing separation from the previous mode of existence), liminal rites (rites symbolizing the transitional stage proper; van gennep 1960: 21), and rites of incorporation (rites symbolizing reintegration into society with a new status). Much research on the phenomenon of passage rituals, in particular initiation rituals, has, of course, taken place since the early twentieth century, both within the general study of religion and more specifically within PCRN (cf. schjødt 2008: 22–57). suffice it to note here, however, that much of this research has focused on social and psychological aspects of initiation, which will not be discussed 1 a research historical account of these rituals, especially in connection with initiation outside as well as inside of the Nordic area, is found in schjødt (2008: 22–57).
Jens Peter Schjødt, Professor of the study of Religion, aarhus 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. 823–851
BREPols
PUBlisHERs
10.1484/M.PCRN-EB.5.116959
824
Jens Peter Schjødt
in this chapter, whereas some important works concerned with structure and symbolism will be briefly mentioned, especially some ‘classical’ works within the field of PCRN.
Passage rituals usually take place when an individual or a group of individuals are to change their religious or societal status, going from a lower to a higher level,2 a transition which is in principle irreversible and which usually involves some modification in physical appearance (physical ‘marks’ or attributes such as weapons or other paraphernalia), together with the acquisition of numinous knowledge, which is ‘secret’ in the sense that those who have not been initiated are not supposed to know it, although in reality they often do.
in the first instance, we can divide rituals of passage as they exist worldwide into two broad groups: on the one hand, those connected with biological changes in the life of individuals; and on the other hand, those that may be celebrated at any time in the life of an individual and are not determined by biology.3
To the biological group we will count rituals linked to birth, puberty, weddings, and death, which are probably found in every society. To the non-biological category we will include rituals performed in connection with entrance into mystery cults, secret societies, and other social settings, such as groups that see themselves as blood brothers, but also into various individual positions in society, such as priests, shamans, kings, and others. in these cases, some socially or religiously important change in individual status are affected.4 in all these instances, we can observe ritual celebrations in religious — and non-religious
— cultures. This, of course, does not mean that all such occasions are ritually celebrated in all cultures, since some cultures lack, for instance, kingship or priests or secret societies. The point here is that when we look for passage rituals in the pre-Christian North, it is wise to focus on transitions of a biological and/or social kind. Therefore, this chapter will do exactly that, although, as we shall see, the sources do not give us information about all of them.
2 The opposite possibility, going from a higher to a lower level, is of course also a possibility, although it is seldom celebrated in any spectacular way. one may, however, think of army officers whose medals and other symbols of rank are torn off in front of the whole regi-ment. Within PCRN we may also think of certain kinds of nið, which will deprive the intended victim of his rank as an honourable person (è21), even if such rituals are often carried out as magic (è26) and therefore often in secrecy.
3 admittedly, these designations are not very precise. for instance, rituals celebrated in connection with marriage are here seen as ‘biological’ but are not ‘biological’ in the same sense as are puberty rituals or funerals.
4 again, it could be argued that these changes all require at least adolescent status, so to some degree there is, also here, a biological aspect in play.
32 – Passage Rituals
825
Sources
in general, no single ritual description, with one exception — namely, the description by the arab diplomat ibn fadlan of the funeral of a Rus chieftain on the Volga in the year 921/22 — is in itself sufficient for us to reconstruct the ritual sequence in any detail.5 However, hints can be found scattered across the various saga categories, not least in the fornaldarsögur (cf. schjødt 1994, 1999a, 2000b, 2003, and 2008); and the structure and symbolism, although in a mythic setting, can be detected in some of the eddic poems as well as in snorri’s Edda (schjødt 2008: 328–78), some of which are quite informative regarding the symbolism that we can expect also to be part of ‘real-world’ performance of rituals.
archaeological sources are extremely important when it comes to burial rituals, which is why these will be treated in their own chapter (è33), whereas it is more difficult from archaeological evidence to depict and reconstruct passage rituals associated with other transitions. That does not mean that archaeological finds of various kinds cannot be meaningfully explained with reference to other types of passage rituals; many finds were probably linked to such rituals, which we must expect to have played an important role in the religious life of the pre-Christian scandinavians. The problem is that it is often difficult to argue for such a connection as more likely than connections to other ritual categories, even when it comes to the pictorial material, such as, for instance, the gold foils (è31). it may well be that, for example, the pictures depicted on the helmet from Torslunda, apparently showing dancing warriors, should be seen in relation to initiatory rituals of warriors, but again we cannot be certain; and even if they are related to such initiations, we cannot know in which way (è24).6
Birth
We do not know much about rituals connected to birth in the pagan period.
from the folklore of later periods, we know that a whole series of various customs were attached to these occasions in which both the mother and the child might be the focal point. This points to two different aspects of the rituals: namely, the integration of the newborn child into society, and the celebration of the new status of the mother (gotfredsen 1956; Näsström 2002b: 70–72).
5 as we shall see below, the forming of blood-brotherhood, although far from as detailed as the funeral in ibn fadlan, can perhaps be said to be another exception.
6 The following is partly a revised version of parts of schjødt (2008: 328–78).
826
Jens Peter Schjødt
The few hints we have for such rituals concern the sprinkling of water ( ausa vatni),7 the giving of a name ( Rígsþula st. 34), and the placing of the newborn on its father’s knee ( knésetja). These should probably be seen as parts of one ritual sequence, the aim of which was to incorporate the child into society (de Vries 1956–57a: i, 179; Näsström 1996b). although we cannot place these three rites within the sequence, it seems likely that the pouring of water could be understood as a rite of separation with a cleansing function, whereas giving the child a name and placing it on the father’s knee should probably be seen as rites of incorporation, which invest the newborn child with the identity that it will have in society.8
it is possible that some form of divination was carried out in association with birth. in Helgakviða Hundingbana I (st. 1–8), set in a mythic scenario, we hear that the norns (è59) arrived after Helgi was born and twisted with strength the threads of destiny (‘snero þær af afli ørlǫgsþátto’, st. 3). The extent to which we can use this particular text as support for general birth rituals is certainly debatable, however, both because the tone of the poem is mythical and because Helgi is by no means an ordinary child: he is a king and destined to become one of the greatest of warriors of all time (see è35).
The sources that we have for these rituals, then, do not allow us to reconstruct a ritual sequence, apart from the help we get from van gennep’s three-phase model (see above) and by using comparative material. on this basis, however, it appears quite certain that such rituals did take place and that they should definitely be viewed as passage rituals of a sort.
Puberty
it is somewhat paradoxical that puberty rituals are hardly represented in the Norse source material, since these particular rituals have been of the greatest interest within the History of Religions and social anthropology in general.
one of the reasons could be that some of the warrior initiations, which we 7 Hávamál 158 suggests that this rite could perhaps also be part of the initiation of a noble ( þegn) and that Óðinn played a part in such cases.
8 There can hardly be any doubt that these rituals played a significant role for the child’s identity as part of society. for instance in Harðar saga ch. 8 we learn that it was murder to kill children who had had water poured over them, whereas before that happens they could be left outside for exposure (cf. Hasenfratz 2011: 63). likewise, name-giving was a serious matter, and the name chosen for an individual often appears to be associated with the religious sphere, even if this can be debated (see è5).
32 – Passage Rituals
827
figure 32.1. a richly furnished grave of a girl about ten years old, at ire in Hellvi on gotland, from the tenth century (grave 218a). in Viking age graves on gotland there are clear patterns associating female objects with the age of the dead. small girls up to five years of age were buried with only a few beads, and in graves together with adults. from about five to fifteen years, girls were buried in graves of their own, with 100 to 250 beads and rings on their right arm. from about fifteen years and up, young women were buried with about twenty beads and rings on their left arm. from this analysis, it seems that girls on gotland went through important passage rituals when they were around five and fifteen years old (Thedéen 2011). after stenberger 1962: 54.
828
Jens Peter Schjødt
figure 32.2. a gold foil figure
with an embracing couple from
Krokek in Östergötland (sHM
21517:111325). This and other
motifs have been understood as
mythological scenes (è43.1),
but they have also been inter-
preted as depictions of different
legal procedures, such as
weddings (Ratke and simek
2006). Photo: gunnel Jansson,
statens Historiska Museum,
stockholm.
have dealt with in some detail above (è24) and will return to briefly later in this chapter, should also be seen as a kind of puberty initiation for boys: becoming an adult male among both the germani and the scandinavians also involved various skills in the use of weapons.9 This lack of distinction, caused by the source situation, is also clearly reflected in two famous monographs on the subject of passage rituals within the germanic area: namely, the works by lily Weiser (1927) and otto Höfler (1934), which we have dealt with rather extensively in (è24). Neither of these authors makes much effort to distinguish between the two categories of puberty and warrior initiations, but this may, as just mentioned, accord with the actual situation as it is represented in 9 of course there must have been significant differences between, for instance, warriors who were to become members of a warrior band and more ‘ordinary’ sons of farmers. We can speculate whether the former category emphasized martial skills much more (e.g., Hálfs saga ch. 10) compared to the latter, which perhaps tended to be a more common ‘puberty’ ritual. as we have seen above (è24), most of the sources we have for reconstructing warrior initiations most likely portray initiations into warrior bands.
32 – Passage Rituals
829
the sources. according, for example, to some sources ( Hálfs saga ok Hálfsrekka ch. 10 and Jómsvíkinga saga ch. 24), the age of those who joined the warrior-bands was between twelve and eighteen. This hardly reflects reality, but it does suggest that even boys who had not reached their teens were seen as capable of bearing arms, and thus seen as warriors, although of a different kind from adult warriors.
from the two sagas just mentioned, we also learn that there were certain rules that the warriors of these bands had to adhere to. They were, for instance, required to be strong enough to lift a large stone from the ground, they must be able to carry certain weapons, and they had to be so tough that they did not need to have their wounds dressed the same day as they were wounded;10 apparently there were also some ethical rules they had to obey (they were not allowed to take children and women as prisoners).
as with birth rituals, we have no descriptions of ritual sequences that might allow us to reconstruct any details in the puberty initiations. Even so, we must undoubtedly accept that these, too, constituted important parts of PCRN.
Weddings
The old Norse word for wedding is brúðlaup, literally meaning ‘bride running’, probably indicating the rush of the participants in the feast,11 although we do not know to what extent this was part of the religious dimension of the ritual.
Regarding weddings, we do not have any sequential description of the ritual that can be connected to pagan times. We do, however, have a few hints.
We are told in Þrymskviða st. 30, when Þórr, dressed up as freyja, is going to be married to the giant Þrymr, that the giant says: ‘Berið inn hamar | brúði at vígia | leggit Miǫllni | í meyjar kné | vígit ocr saman Várar hendi’ (Carry in the hammer to consecrate the bride, and lay down Mjǫllnir in the lap of the maiden, consecrate us together with the hand of Vár). We cannot be certain, of course, that these elements would be part of every wedding, but as it stands, we may conclude that Þórr’s hammar, perhaps in the form of some symbolic figure (cf. the many hammer-shaped amulets found by archaeologists), was used for 10 Here, we can also think of the trials to which signý in Vǫlsunga saga ch. 7 subjects her sons by stitching the cuffs of their kirtles onto their arms, passing the needle through flesh and skin.
11 The underlying form would be brúðhlaup. according to de Vries (1962a: 59, with references), it originally means ‘heimführung der braut und zwar in schnellem lauf ’ (bringing home the bride in a fast run).
830
Jens Peter Schjødt
consecrating the bride by laying it in her lap, an act obviously connected to fertility. This is in accordance with the role of fertility god that is often attributed to Þórr, for instance by adam of Bremen, even if Þórr’s role according to adam seems to be more linked to the fertility of the soil (è41). But in the context of weddings, the hammer should probably not be seen exclusively as a fertility symbol; perhaps it is more likely to constitute some sort of protection against all evil that might ruin the marriage. snorri also tells us about the goddess Vár, who ( Gylfaginning p. 29) listens to agreements made between men and women and who is, thus, associated with marriages and weddings.12 The mention of her hand may be related to a symbolic handshake, which is known from very early on, perhaps even from indo-European times (Wikman 1957: 308).
The phenomenon of entry into marriage has given rise to rituals all across the world (van gennep 1960: 116–45), and entering into marriage is a transition in more than one sense: first, one of the spouses must leave his or her family to join the family of the other. in scandinavian society, which was predominantly patriarchal, it was the woman who left her family to become a member of her husband’s family (cf. de Vries 1956–57a: i, 186), and this in itself would justify a transitional ritual. But a wedding is normally also a matter of beginning a new chapter in the life of both the husband and wife, as they leave behind some of the possibilities (and limitations) — possibly more so for the woman than for the man — that are attached to life as an unmarried person, and in return acquire those that are characteristic of being a couple. The most obvious thing to mention here is the fertility aspect: the ability to beget children and thus continue the family.13
But even if the sources have next to nothing to say about wedding rituals, it is not hard to imagine which elements must have been included: sexual symbolism, perhaps even symbolic (or real) sexual intercourse, probably some exchange of gifts and no doubt much feasting — which may well have included sacrifices — processions and many other phenomena, well-known from the phenomenology of religion. Many of these are also sometimes found in medieval and later Christian traditions and can perhaps be seen as a partial continuation of pagan wedding rituals.
12 Vár is well attested in skaldic kennings. The plural, várar, means ‘oaths’ (cf. de Vries 1962a: 645).
13 We note here the clear connection between the phallic fertility god, freyr, and sacrifices carried out for him in connection with weddings, reported by adam of Bremen ( Gesta Hammaburgensis 4.27).
32 – Passage Rituals
831
Textual Evidence for Rituals Connected to Death and Burial
as was mentioned in the introduction, we have an enormous amount of archaeological material, which informs us about this category of ritual and will be treated in the next chapter (è33). What we shall do here is to mention a few pieces of textual evidence, among others the famous description by ibn fadlan (see also è33 and figures è33), which we shall analyse in some detail, since it is the only description by an eyewitness concerning such a ritual. although burial rituals are undoubtedly those best described within the category of passage rituals, this does not mean that we are able to reconstruct any such thing as a
‘proto-ritual’ that might have been used all over scandinavia. There are considerable lacunae in the sources regarding both the sequence of the events and the meaning and symbolism associated with the ritual. in addition, we must a priori expect considerable differences in the performance of the rituals in regard of both regional and social distinctions (Roesdahl 1987: 177–80; DuBois 1999: 70–72; Price 2012; cf. also schjødt 2009a and è33). furthermore, our most copious source for funeral rituals, ibn fadlan’s description, although very detailed, is also highly problematic, as we shall see below.
apart from ibn fadlan’s description, the written sources do not give us an entire sequence, but in a few eddic and skaldic poems, in snorri’s Edda and in the sagas, we get some scattered pieces of information, which for the most part belong within a specific literary context, but which may nonetheless be expected to shed light on some aspects of rites that were carried out in pagan times (see also è33).
Eddic poems about the gods contain very little of importance, except for a couple of allusions to cremation ( Hávamál st. 71 and 81, Vǫluspá st.
33, Baldrs draumar st. 10 and 11), which do not tell us anything about the actual performance of rituals. Conditions are not much different in the heroic poems. Here, too, there are a couple of references to cremation (first and foremost in Sigurðarkviða in skamma st. 65–69, in connection with the deaths of Brynhildr and sigurðr). in addition, there are references to funeral feasts and the drinking of erfi ( Atlamál st. 75 and Guðrúnarhvǫt st. 8) as well as the burning of slaves in connection with the funeral ( Sigurðarkviða in skamma st. 67
and Guðrúnarkviða in fyrsta end prose). in the prose introduction to Helreið Brynhildar, we learn that there was a carriage on Brynhildr’s funeral pyre in which she was expected to ride to Hel.14
14 This recalls the carriage found in the oseberg grave, cf. Christensen, ingstad, and Myhre (1992: 119–23).
832
Jens Peter Schjødt
We learn a little more from snorri’s Edda. Thus, in the description of Baldr’s cremation, there is a vague hint at a ritual sequence ( Gylfaginning pp. 46–47).
The text speaks of cremation on a ship at sea, which perhaps may be considered an especially honourable form (Uecker 1966: 83–91). first, we hear that Baldr’s ship, Hringhorni, was to be launched, but could not be moved; the gods therefore sent for the giantess Hyrrokkin. she arrived riding a wolf, went up to the boat, and pushed it out into the sea. Then, Baldr’s body was carried onto the ship. on seeing this, his wife Nanna died of grief, and she was also carried onto the pyre, which was then lit. Þórr stood up and consecrated the pyre with his hammer, Mjǫllnir. Here, snorri tells of a peculiar incident: namely, that a dwarf named litr ran in front of Þórr who then kicked him into the fire. further, snorri says that this funeral was attended by all kinds of beings: gods, valkyries, and different groups of giants. finally, we learn that also Baldr’s horse was led onto the pyre to be burnt with him.
it is not easy to determine whether the individual elements mentioned in connection with this cremation are generally applicable to funerals of prominent persons or are particular and only exist in the world of myth. The Hyrrokkin and litr episodes can hardly have had the status of ritual acts,15
while Nanna’s death may be seen in the light of the burning of widows, which is attested in other sources;16 the same is true of the horse being led onto the pyre. No other sources mention that a funeral pyre was to be consecrated, as this text states that Þórr does, but it seems likely that some form or other of sanctification must have occurred, probably with the aim of furthering life and perhaps with an apotropaic purpose (cf. lindow 1997a: 93). Concerning litr, whose significance is completely obscure, it is not possible to provide a plausible cultic parallel.
15 sune lindqvist suggests, rather speculatively (1921: 171), that Hyrrokkin corresponds to the wind, which by ibn fadlan is said to carry the dead chieftain up to his master (see below).
if we are to interpret the role of this giantess, it must be that she represents fire (cf. the first syllable hyrr = fire), but this is uncertain. for a good overview on this subject and possible parallels to Hyrrokkin, see lindow (1997a: 74–80). lindow’s understanding of the figure (1997a: 88) — that she is a representative of the socially inferior group of giants, but is nevertheless necessary for this, the most serious situation in the life of the gods — seems probable. it is, however, a purely mythological construction, which obtains meaning and significance only with reference to Baldr’s cremation and not to rituals in the human world.
16 Cf. above with reference to Brynhildr and women who seek death with their husbands, as we see in Gesta Danorum 1.8.4 and 2.5.5, and also the arabic traveller ibn Rustah (Birkeland 1954:17).
32 – Passage Rituals
833
Related to the funeral rituals are the memorial feasts, but these should probably not be considered part of the ritual sequence that aims to ensure the dead person’s transition to the world of the dead. instead, they may be seen as an element in the cult of the dead, the function of which was, above all, to commemorate the dead (cf. the noun minni ‘memory, memorial, memorial toast’) and honour them, while transferring the dead person’s property and position to the nearest kin.17
although the description of Baldr’s funeral is the most extensive within the mythological corpus concerning funerary rituals, it is clear that it cannot be used to afford a more detailed insight into the ritual symbolism and structure.18
The other written sources do not contribute much either. although there are several passages that tell us about things taking place when a person has died, we are not in a position to reconstruct a full ritual sequence.19 Even so, the existence of a series of funerary rites is beyond doubt. first of all, we notice 17 There are several examples in the saga material that indicate that the aim of memorial feasts was primarily to transfer to the living not only the possessions of the dead, but also the position of the dead king or chieftain. Thus in Hervarar saga ok Heiðreks konungs ch. 11, we hear that angantýr is crowned king at the thing, but he does not take possession of his high seat until after he has avenged his father, nor does he celebrate the memorial feast until then.
The act of taking possession of the high seat obviously played a considerable symbolic role in the assumption of power and may, therefore, correctly be considered as a reintegration rite in the ritual sequence, with the ‘new leader’ as its object. This compares to Ynglinga saga ch. 36, where it is said that the new king may only sit down in the high seat once he has drunk from the ‘Bragi-cup’ ( Bragafull; on this see de Vries 1956–57a: i, 457–58); however, when this has taken place he can enter fully and completely into his new role. Emil Birkeli (1944: 19–21) has clearly demonstrated that the high seat had both a social and a religious significance. from the passages mentioned, it also appears that, in connection with drinking, oaths are sworn (cf. è31) with regard to various forms of heroic action that will be carried out in the future, thereby marking the qualitative difference that the new status will bring. Whether an erfikvæði (literally: ‘inheritance poem’) was recited during the funeral feast itself cannot be determined with any certainty, but a passage in Egils saga Skallagrímssonar ch. 78 may indicate this. it is obvious that the arrangement of an erfi, in addition to the function it may have had with regard to the heir, also served to honour the memory of the dead person. such an erfi, however, should probably not be seen as part of bringing the dead into the other World.
18 Húsdrápa’s description of Baldr’s cremation does not add anything new on the subject of mortuary ritual to snorri’s statement, except perhaps that some kind of procession was involved. That processions formed part of funeral rituals in the real world is certainly what should be expected (see also Nygaard and Murphy 2017: 51–65).
19 an attempt to set up a sequence of events, based on saga material, has been carried out by inger Vibeke Hansen (1981: 135–39); but also here it appears that this can only be done with a very high degree of uncertainty.
834
Jens Peter Schjødt
that some of the acts described in various sources must have served an apotropaic purpose, such as closing the dead person’s eyes (cf. ström 1942: 242–
44), and such acts would not necessarily have been part of the ritual sequence in which the passage from one world to another was the object. However, an interesting rite is described in several sources: namely the ‘preliminary burial’
(e.g., Droplaugarsona saga ch. 6 and Brennu-Njáls saga ch. 17), to which we shall return below in connection with ibn fadlan’s description. further, in all instances it is usual for the corpse to be covered, which is perhaps due to the risk of being molested by wild animals, as is indicated in Grágás (i, 88). People, moreover, gave burial gifts of various kinds, both animal and human, which the archaeological material also demonstrates; how this was done is, however, far from clear. it likewise appears from several passages that, prior to the burial itself, the corpse had to be prepared according to some customs (e.g., Gísla saga ch. 14). This may have included the closing of the nostrils and eyes, together with the dressing of the corpse, and possibly the binding of shoes onto the dead person’s feet ( helskór; Gísla saga ch. 14 and 17).20 However, it is hardly possible here, either, to discern a certain symbolic pattern, although it is likely that different rites of separation were involved. Evidence for liminal rites are almost nonexistent. Most interesting, perhaps, is the statement by snorri in Hákonar saga góða ch. 32, where it is said in connection with Hákon’s death that his grave was spoken over as was customary among the pagans, and that this in some way ‘vísuðu honum til Valhallar’ (showed him the way to Valhal). What was said is unlikely to be the poem Hákonarmál as we know it, although snorri quotes it at the end of the chapter, since the wording used by snorri rather suggests the kind of formulas we know from other religions21 and which are intended to lead the dead person to the afterworld, thus making it easier for 20 as this passage is the only place where the binding of ‘ hel-shoes’ is mentioned, it may perhaps be doubted whether this was a common custom. one can argue, however, as is done by Helge Rosén (1919: 128–29), that the custom was very common, since the author of the saga presupposed that his readers would know it. Whether Rosén, and several others who also accept this piece of information as authentic, are correct in presuming that the custom was in general use in scandinavia, we cannot know for certain, but it seems reasonable to argue in this case, as in others, that customs that were known by most people would not be mentioned each time they took place. Thus, argumenta ex silentio are useless here. Rosén’s argument is tenable as far as it goes, so that in some environments in iceland at least the custom would have been generally known.
21 The closest examples are the so-called orphic gold tablets, found in graves in greece and italy, which instruct the dead on what to do and what not to do at the various stages on the road to the kingdom of the dead (see Zuntz 1971: 299–393; albinus 2000: 141–52).
32 – Passage Rituals
835
him or her to enter it. Here, we may once again link to Baldr’s funeral, where snorri furthermore says that Óðinn whispered something into the ear of his dead son. The poem Hákonarmál, however, seems not in itself to have the character of instructions about how to behave in the hereafter and must be regarded primarily as a poem of homage.
some people might also on their death-bed have been marked with a spear, as is said about Óðinn and Njǫrðr in Ynglinga saga ch. 9. The text says directly that this marking will secure for the dead a life with Óðinn, probably in Valhǫll, and we may therefore speculate whether this custom was specific for odinic warriors who were not killed in battle (cf. schjødt 2007a). But again, this is mentioned only in a mythic-legendary context, so we cannot be sure of what status this action had in the socio-historical context.22
The Description by Ahmad Ibn Fadlan
The only pre-Christian ritual that is described as an entire sequence is the funeral of a Rus chieftain somewhere on the river Volga, by the arab traveller and diplomat ahmad ibn fadlan, whom we have already discussed earlier (è23) and who will be mentioned throughout this work (in particular in è33), in his small book Risalat. 23 ibn fadlan was a member of a diplomatic mission instigated by the caliph al-Muktadir during the years 921–22 to the Volga Bulghars, and somewhere along the Volga he met with a group of Rus, most likely swedish ‘Vikings’.24 The book deals with various peoples and gives important information about the lifestyle of the Rus, but the important passage, which we shall summarize briefly here, is the description of a funeral of one of their chieftains. Being an eyewitness account, ibn fadlan’s description is outstanding and carries much greater value than any other ritual description, both because it is much more reliable and because it is much more detailed.
22 a few other relevant old Norse texts, but also the funeral scenes in Beowulf, will be mentioned below, in particular, Ynglinga saga ch. 8 concerning Óðinn’s decisions concerning rules for burials, will be treated below in (è33).
23 Part of the information related by ibn fadlan is also mentioned by other arab writers (see Thorir Jonsson Hraundal 2013: 120–21). for other relevant written sources including the Byzantine, see (è33).
24 for a good overview of the history of the Rus and many problems connected to their identity, see Thorir Jonsson Hraundal (2013: in particular 123–28) and Duczko (2004; cf. also è15). Duczko gives a translation of the relevant passage in (2004: 139–41). another, more philologically oriented translation, is James E. Montgomery’s (2000) with many valuable commentaries (cf. also Montgomery 2008).
836
Jens Peter Schjødt
There are problems, however, due to several factors. first and foremost, ibn fadlan was an outsider and thus dependent on a translator of whom we know nothing, such as whether he was Rus or perhaps another Muslim; and it seems that there were elements in the ritual that he did not really understand. He also
‘translated’ some of the notions into arabic, which makes it difficult to find our way ‘back’ to the scandinavian concepts. for instance, he quotes the slave woman (see below) as saying that she looks into ‘paradise’, a term that she would almost certainly not have used, but we do not know whether she perhaps used a word designating Valhǫll. another problem is that we do not know to which extent the group of Rus described were travellers, were staying by the river for a shorter period, or if they had perhaps been there for generations. if the latter is correct, we must assume that they were heavily influenced by slavic and other cultures (the Volga Bulghars were, for example, a Turkish-speaking group; cf. Thorir Jonsson Hraundal 2013: 112), including customs related to religion; and if so, it is obvious that it will be hard to use the description as a means to reconstruct funeral rituals in scandinavia during the same period. for these reasons, ibn fadlan’s text is problematic as a source, although the problems are of another kind than those we usually envisage when we deal with PCRN. Having said this, however, we can also state that there are several elements in the text that correspond to the scandinavian context without problems.
ibn fadlan starts out by speaking about some general customs connected to death: for instance, that deceased individuals (apparently both rich and poor) are placed in boats, and the property of a rich person who has died is divided.
We are further informed that when someone has died, his family will ask the servants who among them will die with him, and the volunteer (?), most often a girl, cannot take back his or her word, even if he or she should later wish to do so. ibn fadlan has been told that a prominent Rus man has died, and the rest of the description is an account of what is done in connection with this death. first, we hear that the deceased was put in a provisional grave for ten days,25 and that a slave woman who is to follow her master onto the pyre is then selected. During the rest of the period, she is accompanied by a couple of girls (who are the daughters of the ‘angel of Death’, whom we hear about later), who even wash her feet. This can only mean that the girl has now changed her status and should be seen as the chieftain’s wife, at least symbolically.26 When the 25 These ten days should probably be seen as a period of nine nights and thus as a kind of
‘sacred’ period.
26 This is confirmed in the version by amin Razi where the woman is said to be the wife of the deceased (cf. Duczko 2004: 145).
32 – Passage Rituals
837
day of the cremation arrives, the deceased chieftain’s ship has been pulled up onto the beach and is supported by four planks of birch, while other preparations have also been made. Then people begin to walk around the boat, speaking words that ibn fadlan states that he does not understand.27 after that, we hear about the ship being made ready for the chieftain to be placed in. This was done by a figure called ‘The angel of Death’, whom ibn fadlan sees as ‘a strap-ping old witch’ (Duczko 2004: 139), who is said to be responsible for some of the ‘practical’ tasks during the funeral, although some of these tasks are clearly religious, such as the killing of the servant girl. in this way, she clearly acts as a kind of priestess. Then, some of the Rus go into the provisional grave in order to bring the dead man from there to the ship, and we learn now that they had placed some fruit and alcohol in the provisional grave. They dress him for the cremation and place alongside him various kinds of vegetables and musical instruments and put him into a tent (or a pavilion, cf. Montgomery 2000: 16), which is placed on the ship.
ibn fadlan then goes on to mention the various animals that are sacrificed: horses, cows, poultry, and a dog, which are all killed in specific ways. for instance, we are told that the horses were made to gallop until they began to sweat and were then cut up and thrown into the ship; the dog, however, is cut in two and also thrown into the ship.
Meanwhile the servant girl, drinking and singing, goes into one tent after another, visiting the friends of the deceased, and in each case she has sexual intercourse with the owner of the tent who then asks her to tell her master that he does this for love of him. later on in the afternoon, they bring the girl to a construction resembling a doorframe. There, she places her feet on the hands of the men, and when they raise her above the doorframe, she utters some words. This is repeated three times, and then she is handed a hen, the head of which she cuts off and throws away, while the men throw the hen into the ship.
Having asked the interpreter what is going on, ibn fadlan is told that the first time they lifted her, she said: ‘Behold, i see my father and mother’.28 The second time, she said: ‘Behold, i see all of my dead kindred seated’, and the third time:
‘Behold, i see my master, seated in Paradise. Paradise is beautiful and verdant.
27 since he did not understand the language of the Rus, it can be assumed, perhaps, that what was spoken in this situation was different from ordinary speech, since it is explicitly mentioned that he did not understand it, and perhaps we are here dealing with some sort of ‘ritual language’.
28 This is also an indication of the status shift mentioned above. The parents of a slave or servant would hardly be conceived of as inhabiting the world to which the chieftain is about to go.
838
Jens Peter Schjødt
He is accompanied by his men and his male slaves. He summons me, so bring me to him’ (p. 18). Then, she is brought to the ship where she removes her jewellery and gives it to the angel of Death and her two daughters, who are identical to her two guards mentioned earlier. she is given more alcohol and sings in order to say farewell to her friends. The girl and the angel of Death then go into the tent on the ship and six men follow, who all have intercourse with her. after that, a rope is placed around her neck and is pulled by two of the men, while two others take hold of her feet and two of her hands. The angel of Death then approaches with a dagger, thrusting it several times in between her ribs, while the two men strangle her with the rope until she dies.29
afterwards, the closest relative of the deceased lights a piece of wood and approaches the pyre walking backwards, facing the spectators, and being completely naked. He has the lighted piece of wood in one hand while the other covers his anus (cf. sass and Warmind 1989), which should perhaps be seen as an apotropaic act. Then, he ignites the wood, and the other people come forward with firewood and burning sticks, which they throw onto the pyre. a hard wind, then, begins to blow, and within an hour everything is burned. one of the Rus says that this wind was sent by their god. finally, a round hillock is built on top of the ashes where the ship had been, and on that they erect a piece of wood on which is written the name of the chieftain and the name of the Rus king.
as can be seen, we do get a lot of information in this extraordinary description, and although this is not the place to analyse all the details, some remarks are appropriate.30 first of all, we notice that the full ritual sequence involves not one but three ‘passages’: first, we have the dead chieftain who is thought to go from this world to the other World; second, we have the slave girl, whose status is changed and who also later has to leave this world and accompany her
‘husband’ into the other World; and third, it appears that we glimpse part of an inauguration ritual, in which the closest relative, perhaps a son of the deceased, lights the pyre. The way this is done recalls some elements that we often meet in initiation rituals (cf. Eliade 1975), in which a new leader comes into office: namely, the fact that he is naked and that he walks backwards towards the pyre.
Therefore, he faces the group (the living) whose leader he is going to be, and he approaches them from the other World and thus as a ‘newborn’ or ‘reborn’
leader, since nakedness often functions as a symbolic characteristic of the neophytes in initiation rituals (Eliade 1975: 32, passim). ibn fadlan was much 29 This combination of strangling and stabbing has reminded many scholars of Óðinn’s self-hanging (è42).
30 a more detailed analysis is found in schjødt (2007b; 2008: 344–52).
32 – Passage Rituals
839
more focused on what happened to the dead chieftain and the slave girl, and it is quite likely that there could have been several rites, of which we are told nothing, whose object was the new chieftain. Perhaps they were carried out in secrecy or at least less publically.
This distinction between three ritual sequences for the three main subjects involved is, of course, only an analytical move. ibn fadlan does not distinguish between them, nor do most historians of religion. Nevertheless, it can be useful to bear in mind that a funeral ritual is not always only a funeral ritual, although all the various rites are, obviously, triggered by a death.
Perhaps the most important merit of ibn fadlan’s description is that it allows us to see a ritual sequence. from the death of the chieftain to the final raising of the hillock it is actually possible to apply van gennep’s three-phase structure: The preparations taking place while the chieftain is in the provisional grave and the appointing of the slave girl, perhaps together with many other rites of which ibn fadlan was not aware, constitute the separation phase. Most of the rites carried out on the day when the chieftain is burned can be seen as liminal. The exception are the rites performed after the burning, such as the raising of the hillock and the erection of the piece of wood with the name of the Rus king, which classify as rites of incorporation: The chieftain has now been transferred to the other World, and the living will return to their ordinary mode of existence. as is evident, there is much sexual symbolism involved, and this is probably an important feature in bringing the dead into a new mode of existence, since the men do it for love of their ‘master’. There is also a great deal of drinking going on, especially in connection with the slave girl, who appears to be drunk most of the time, at least on the day of the funeral. This can easily be linked to some of the ritual feasts already mentioned, where drinking clearly has ritual properties. other ‘classic’ religious features are the walking around the pyre, a kind of circumambulation, and not least the sacrifices that take place. There can hardly be any doubt that the way these animals were killed also had a special symbolic value (the horses and the dog in particular), but we cannot know the significance of all the individual rites mentioned by ibn fadlan. This leaves us in a situation that necessitates much interpretation.
an important question, mentioned above, is to what extent this ritual description reflects Viking age scandinavian customs. Even though we cannot be certain, there are a number of elements that suggest a rather close resemblance: for instance, the words spoken by the slave girl when she is lifted above the doorframe and looking into ‘Paradise’. she sees her dead relatives and her master surrounded by his men, which may well be a vision of Valhǫll. if a Muslim had to find a word for this very positively connoted place, he would
840
Jens Peter Schjødt
hardly hesitate to call it ‘Paradise’. another example is the god who sends the wind in order to take the chieftain to this otherworldly abode. if we consider which god is likely to invite chieftains to his kingdom, and who among the gods rules over the wind, the answer is most certainly Óðinn. in Ynglinga saga ch. 7, he is said to be able to manipulate the wind with mere words,31 and he is the patron god of noble men who will go to Valhǫll when they die (schjødt 2007a). This combination of a god of the dead, having a specific relation to chieftains and other noblemen, and also god of the wind may not be particular for scandinavians, but it could very well suggest that Óðinn was the god of Rus chieftains. finally, a minor detail seems to be decisive when it comes to determining whether the Rus ritual links to scandinavian religious ideas. after the slave girl is lifted up above the doorframe, she takes a hen and cuts off its head, which she then throws away, whereas the body of the hen is thrown onto the ship. a clear parallel to this, although in inverted form and in a completely different context, is found in Gesta Danorum 1.8.14.32 it is part of the story of Hadingus, which relates how the protagonist was invited by a woman from the underworld, which is of course also the world of the dead, to join her in order to see where he would go after his death, as saxo philosophizes. Having passed through a fog, he sees a sunny place with fresh grass. Then they arrive at a river in which various weapons are flowing and across which there is a bridge.
once they cross, they see two armies ( acies) fighting.33 These are people, the woman explains, who have died in battle and who are doing the same as they did while they were alive. Then they come to a wall, which is difficult to cross, but the woman cuts the head off a rooster and throws it over the wall. The rooster crows from the other side, proving that it is now alive again. Without further information, we are then told that Hadingus is back in his own world.
as just mentioned, there are important differences between the two episodes when it comes to context. ibn fadlan describes a ritual act, whereas saxo deals with a mythic or semi-mythic episode. However, it is not surprising that 31 This may seem to contradict the idea of Þórr as the wind raiser (Perkins 2001; cf. è41).
However, we do not have to see each individual function as belonging only to one individual god (cf. schjødt 2012b). if Þórr is really a wind-raiser, he is so by a physical act — namely, by blow-ing — whereas Óðinn manipulates the wind, as snorri says, by words and thus by a mental act.
32 Duczko (2004: 149) has also noticed this parallel, but he does not take into consideration the differences between the two incidents described by saxo and ibn fadlan, respectively.
Thus, he argues that the hen may actually have been a rooster, ignoring the fact that the two incidents constitute an inversion of each other (see below).
33 The description with two armies, consisting of men who have died in battle, of course recalls the various descriptions of Valhǫll (è34).
32 – Passage Rituals
841
symbolic structures found in myths can also be found in rituals. other differences, however, should rather be seen as inversions: ibn fadlan refers to a hen, whereas saxo mentions a rooster. The slave girl and Hadingus, although both deal with the world of the dead, approach it from different directions: Hadingus is in the realm of the dead, whereas the slave girl is still among the living. This means that the woman in saxo throws the head into the world of the living, where it crows as a sign of life, whereas in ibn fadlan’s description, the body of the hen is thrown into the world of the dead, that is, onto the ship.
Thus, there is a series of inversions: in ibn fadlan, a woman is about to join the dead, a hen is decapitated, and its body thrown into the world of the dead by some men; whereas Hadingus is about to join the living, a rooster is decapitated, and its head thrown into the world of the living by a woman.
all this means that, although the two episodes are very different, they are so in a very systematic way. The elements are the same: a person about to go from one world to another, a decapitated poultry bird, the head remaining in or thrown into the world of the living, or the body remaining in or thrown into the world of the dead. The parallels play on a range of oppositions: masculine vs. feminine, head vs. body, this world vs. the other World, life vs. death. This structural parallel, although in the form of an inversion, is so detailed that even if we are not ready to accept all of the other parallels noted above and elsewhere, it cannot be explained away by pointing to general features in the religions of the world or to coincidence. Therefore, it seems indisputable that these Rus people and their world-view were at least to some extent scandinavian-derived. We will obviously never know whether the other elements might likewise be typical scandinavian or whether they constitute a mixture of traits from multiple cultures, as suggested by Thorir Jonsson Hraundal (2013: 128), but it would be wise not to reject the information presented by ibn fadlan, for instance, in the interpretation of archaeological finds, as has also been accepted by many archaeologists (e.g., Price 2012; è33).
Initiations into Warrior Bands
We have already dealt with initiation into the warrior bands (è24) in some detail and will therefore be brief in this section. Unlike the transition from the living to the dead in the funeral rituals, we are not very well informed about warrior initiations, nor, as mentioned, about most other passage rituals. in fact, we only have some semi-mythical accounts, mostly from the fornaldarsögur.
We have treated the sigmundr-sinfjǫtli episode from Vǫlsunga saga and the Bǫðvarr-Hjalti episode from Hrólfs saga kraka above (è24), both of which
842
Jens Peter Schjødt
seem to be reminiscences of rituals that are likely to have been performed in connection with initiations into warrior bands. in these sagas, a few more incidents are described, which may be part of the same ritual complex, but none that add substantially to our knowledge of such initiations.
What can be said with some certainty is that such rituals would have involved tests of skills and courage, as is obvious in the two saga episodes just mentioned, but also for instance Hálfs saga ok Hálfsrekka ch. 10 and Jómsvikinga saga ch. 6 — both mentioned above in connection with puberty rituals — yield information about what was needed to become accepted into these warrior bands. Even so, we are nowhere explicitly told about religious rituals in that connection. However, another important feature is the link to Óðinn, pointed out already by otto Höfler (1934: 323–41) who at the time termed the institution of the warrior band a Weihekriegertum, thus clearly indicating the initiation aspect. Thus, it appears rather unproblematic to infer initiation rituals, most likely associated with Óðinn (cf. schjødt 2008), whenever these bands are mentioned. further, on the basis of these semi-mythical sources, we moreover obtain the idea that, apart from the tests, some sort of symbolic transformation into animals was also involved, mainly wolves and bears, but perhaps also boars (cf. Hedeager 2011: 89–90), and other animals. it seems clear that these warriors were connected to Óðinn, not only when alive but also in death (schjødt 2007a), becoming the einherjar of Valhǫll, the warriors who will fight the giants at Ragnarǫk and until then will carry on happily drinking and fighting. We cannot reconstruct a detailed ritual sequence from these sources, but it appears that they followed the three-phase scheme outlined by arnold van gennep: first, a separation from the ordinary society by performing the tests, which would make the neophytes ready for the second step, which is the liminal period during which they were probably identified with certain odinic animals and perhaps going through a symbolic death (cf. sinfjǫtli, è24). finally, we have the third step, when they were reintegrated into society and received a new name and some warrior equipment.
Even if the sources appear meager, we can be confident that, just as the institution of the warrior band played a significant role among the germani right up until the Christianization (and perhaps even after), the accompanying initiation rituals formed an important part of the religious and social life of the military elite.34
34 in an intriguing article, Michael Enright (2007) has proposed an interpretation of one of the plates (iV) of the gundestrup cauldron as depicting a warrior initiation. However, since the cauldron is of Celtic or Thracian provenance, it cannot be taken as direct evidence of initiations into the germanic/Nordic warrior bands.
32 – Passage Rituals
843
Initiation of Rulers
as with the initiation rituals of warriors, we have almost only indirect sources pertaining to the initiation of kings and other rulers, which have likewise been dealt with above (è23). There, we noticed that it seems as if the new king had to be placed in a high seat as part of the inauguration, confirming that at least some symbolic actions took place when a new ruler came into power, as has also been argued by a number of scholars (e.g., sundqvist 2001, 2002, 2016). in this brief section, therefore, we shall only deal with one aspect of these rituals: namely, the transmission of numinous knowledge, which is part of most initiation rituals across the world.
figure 32.3. ‘Mora stenar’ according to olaus Magnus in Historia de gentibus septentrionalibus, a history of the Nordic people, printed in Rome in 1555. The stones at Mora in central Uppland were the site where swedish kings were elected in the Middle ages. at the election, the king stood on a large stone slab. These rituals concerned the inauguration of Christian kings, but the ritual element of standing on a large stone slab may be older, and may have been an alternative to initiating a ruler in a high seat (Zachrisson 2010). after: Magnus 2010 [1555]: 324.
in an interesting article from 1970, Jere fleck proposes that some of the eddic poems ( Hyndluljóð, Grímnismál, and Rígsþula) were concerned with such transmission. in Hyndluljóð, the frame-story is that freyja is riding to the underworld on a boar, which in reality is Óttarr, a king-to-be and her protégé, in order to obtain information from a wise giantess, Hyndla. The main part of the poem is a long series of pieces of information, genealogical as well as mythic, given reluctantly by Hyndla (è45).35 in Grímnismál, the frame-story 35 for an analysis of the initiation theme in Hyndluljóð, see schjødt (2008: 251–61; and more briefly, 2016: 138–44).
844
Jens Peter Schjødt
tells of how Óðinn gave secret information to geirrøðr, his protégé, who later becomes a famous king; subsequently, Óðinn similarly related information to his son, agnarr, who must be assumed to become king after the death of his father.36 finally, Rígsþula tells the story of a god, Rígr, who in the introductory prose is identified with Heimdallr but could also be seen as a manifestation of Óðinn (cf. Cöllen 2011: 82–92, with references; schjødt 2017b), who travels through the human world. By having sexual intercourse with three women from three different strata within society, Rígr becomes progenitor of three social classes: slaves, free farmers, and the nobility. The son of the third woman, who is noble, is called Jarl and his youngest son, in turn, is Konr who is called ‘Konr ungr’, a name clearly intended to imply the noun konungr,
‘king’. from stanza 43 onwards, we are told about Konr’s skills, including knowledge of runes and other sorts of magic, and as it happens he becomes more knowledgeable about runes than his father, thereby acquiring the right to call himself Rígr (st. 45) (è50).37 Through a comparison and analysis of these three poems, fleck finds that all three portray a divine actor impart-ing knowledge to a potential king, and that the relationship between god and protégé is crucial for the discharging of the latter’s duties in the sacral kingship. Thus, for fleck, the only reasonable way to understand Rígsþula st. 45, where Jarl and Konr are discussing runes, is that Konr proves superior because Rígr has instructed him. fleck’s conclusion to his analysis of Rígsþula is that education in numinous knowledge, supplied by some divine figure, formed a decisive factor in the succession to a germanic sacred kingship (fleck 1970: 42).38 This idea seems to be confirmed when we look at other ‘kings’ in the semi-mythic texts, such as sigmundr and sigurðr in Vǫlsunga saga, Hadingus and Haraldus Hyldetan in Gesta Danorum, and others. They all have links to Óðinn, and they all receive advice and other gifts from him (è42). Thus, even if pagan inauguration rituals, at least those elements that were directly associated with pagan gods, had to be abandoned with the Christianization, 36 Whether or not Óðinn’s visions in Grímnismál should be regarded as information intended for agnarr can, however, be discussed (see schjødt 1988 and Nygaard 2019) 37 Rígr is derived from irish rí, meaning ‘king’; accordingly, it should here be seen as a title.
38 fleck also argues that the fact that both Konr and geirrøðr are youngest sons is important, because it shows that ideally younger sons would become kings instead of the oldest, if they possessed sufficient knowledge. if this idea is correct, it contributes to emphasize the importance of such knowledge. for further discussion of the last stanzas in Rígsþula, see also von see and others (2000: 646–65).
32 – Passage Rituals
845
it seems possible to get a glimpse of what they were like through these semi-mythic or legendary texts.39
although none of the sources individually considered can be said to constitute ‘proof ’ that such inauguration rituals were performed, when we take all the evidence into consideration, it seems that we can be confident that they actually were, indeed, performed. Not least adopting a comparative perspective strongly suggests this, since there is hardly a single example of new kings whose ascen-sion to that title has not taken them through such ceremonies; and in religious societies, these are always more or less directly connected to the other World.
The Formation of Blood-Brotherhood40
as opposed to initiations of kings and elite warriors, which we find in almost all societies, there also existed an institution in pre-Christian scandinavia, which is only found in some societies. This is the blood-brotherhood: the mixing of blood among a number of men, who thereby create a kind of artificial family relation, which was, however, just as strong as a ‘real’ one and included obligations to avenge the killing of one’s blood-brothers.41 We may speculate whether this goes back to some sort of military fraternities, and should therefore perhaps belong in the section on warrior initiations (cf. Hellmuth 1975: 202–18). However, as we meet the institution in the medieval sources, a direct connection cannot be discerned. The institution of blood-brotherhood is mentioned in several textual passages,42 but it has left no traces in archaeology.
Paradoxically, we see that even if the institution apparently does not hold the same amount of public importance as, for instance, the institution of kingship or warrior bands, we are much better informed about the initiation into it.
several reasons for this could be suggested, one of them being the absence of 39 as has been argued by gunnell (1995 and many others), it is quite likely that some of the eddic poems were parts of rituals. if this is also the case with Rígþula, we can easily imagine that the poem would be part of a ritual aiming at supplying the new king with numinous power by a human figure impersonating the god (whether Heimdallr or Óðinn).
40 This section is an abbreviated and revised version of schjødt (2008: 355–73).
41 of course, this relation would be different from that of the biological family, in the sense that the blood-brotherhood did not involve obligations of any sort to family members of the other blood-brothers, including vengeance. Therefore, the blood-brotherhood was a kind of
‘closed’ community, only involving the members, and thus reminding us of the members of the warrior bands.
42 Hellmuth (1975: 14–54) contains a complete register of the passages in the sources.
846
Jens Peter Schjødt
warrior bands and kings in iceland, whereas blood-brotherhood was apparently quite common.
The main sources for the rituals initiating a man into blood-brotherhood are three saga descriptions and a passage from Gesta Danorum. They are all presented here in full.
Gísla saga Súrssonar [ch. 6]
ganga nú út í Eyrarhválsodda ok rísta þar upp ór jǫrðu jarðarmen svá at báðir endar váru fastir í jǫrðu ok settu þar undir málaspjót þat er maðr mátti taka hendi sinni til geirnagla. Þeir skyldu þar fjórir undir ganga, Þorgrímr, gísli, Þorkell ok Vésteinn.
ok nú vekja þeir sér blóð ok láta renna saman dreyra sinn í þeiri moldu er upp var skorin undan jarðarmeninu ok hrœra saman allt, moldina ok blóðit. En síðan fellu þeir allir á kné ok sverja þann eið at hverr skal annars hefna sem bróður síns ok nefna ǫll goðin í vitni. (pp. 22–23)
(They walked out to Eyrarhvolsoddi and scored out a long strip of turf, making sure that both ends were still attached to the ground. Then they propped up the arch of raised turf with a damascened spear so long-shafted that a man could stretch out his arm and touch the rivets. all four of them had to go under it, Thorgrim, gisli, Thorkel and Vestein. Then they drew blood and let it drip down onto the soil beneath the turf strip and stirred it together — the soil and the blood. Then they all fell to their knees and swore an oath that each would avenge the other as if they were brothers, and they called on all the gods as their witnesses.) (p. 7) Fóstbrœðra saga [ch. 2]
Hafði sú siðvenja verit hǫfð frægra manna, þeira er þat lǫgmál settu sín í milli, at sá skyldi annars hefna, er lengr lifði, þá skyldu þeir ganga undir þrjú jarðarmen, ok var þat eiðr þeira. sá leikr var á þá lund, at rísta skyldi þrjár torfur ór jǫrðu langar; þeira endar skyldu allir fastir í jǫrðu ok heimta upp lykkjurnar, svá at men mætti ganga undir. (p. 125)
(it had been a tradition among men of renown to become bound to each other by a law which stated that whoever outlived the other would undertake to avenge his death. They had to walk underneath a triple arch of raised turf, and this signified their oath. The arch was made by scoring out three lengths of turf and leaving them attached to the ground at both ends, then raising them to a height whereby it was possible to walk underneath them.) (p. 331)
Þorsteins saga Víkingssonar [ch. 21]
‘Nú vil ek bjóða þér þann kost, ef þú gefr Bela líf, at vit sverjumst í fóstbrœðralag’[…]
var þettta síðan bundit fastmælum. Þeir vöktu sér blóð í lófum ok gengu undir jarðarmen ok sóru þar eiða, at hverr skyldi annars hefna, ef nokkurr þeira yrði með vápnum veginn. (p. 234)
32 – Passage Rituals
847
(‘Now i will offer you that condition, if you let Beli live, that we swear foster-brotherhood’ […] This was later confirmed by a firm engagement. They let blood run into the palms of their hands and went under a strip of turf and there swore oaths that each would avenge the other if one of them was killed by weapons.) Gesta Danorum [1.6.7]
spoliatum nutrice Hadingum grandeuus forte quidam altero orbus oculo solitarium miseratus lisero cuidam pirate solenni pactionis iure consiliat. siquidem icturi foedus ueteres uestigia sua mutui sanguinis aspersione perfundere consueuerant, amicitiarum pignus alterni cruoris commercio firmaturi. (p. 48)
(an aged man with only one eye happened to take pity on the lonely Hading, robbed of his nurse, and brought him into friendship with a pirate liser by establishing a covenant between them. Now our ancestors, when they meant to strike a pact, would sprinkle their combined blood in their footprints and mingle it, so as to strengthen the pledge of their fellowship.) (p. 49)
There are a few other hints to this ritual, especially about blood running into footprints,43 but, as in saxo, without the so-called jarðarmen rite being mentioned.
although all these sources are of a fairly recent date in relation to the time that they pretend to describe, most scholars agree that they do, in fact, give us some idea of how the formation of blood-brotherhood took place, not least because the customs are so ‘strange’ that an influence from Christianity is very unlikely. However, there is considerable disagreement about the symbolism of the ritual(s) described.44 Common for the first three sources is that the participants are to ganga undir jarðarmen (go under a strip of turf or an arch of raised turf ),45 and that they must swear an oath; the two actions appear in Fóstbrœðra saga to be identical — going under the turf is the oath. in all three saga episodes, the oaths are closely connected to vengeance: he who lives longest must avenge the other(s). Thus, the ritual will have consequences for the parties involved for the rest of their lives, and the blood-brothers are henceforth to be seen as actual brothers, as is most obvious in Gísla saga. Maybe we should even speak of a kind of identity among blood-brothers, as was postulated by Jan de Vries almost a hundred years ago (de Vries 1929a: 116) — at least we know that it was thought 43 for instance in Brot af Sigurðarkviðu st. 17, where Brynhildr says: ‘mantattu, gunnarr, til gorva þat, er þit blóði í spor báðir rendot’ (‘you clearly did not remember, gunnarr, that you both let your blood run into a trench’; a more literal translation, however, would be: ‘Do you not remember, gunnarr, that you both let your blood run into your footprints’).
44 a detailed overview is found in Hellmuth (1975: 78–84).
45 literally, jarðarmen means ‘earth necklace’.
848
Jens Peter Schjødt
that important qualities could be transferred from one being to another along with the blood (see, for instance, Hrólfs saga kraka ch. 35). it is therefore a matter of entering a new existence, which is so often part of the symbolism within initiation rituals and passage rituals in general (Hellmuth 1975: 190–92). Here, this is indicated not least by the jarðarmen rite, which most likely symbolizes a kind of rebirth (Pappenheim 1919: 78–80; de Vries 1929a: 132).
The texts quoted above undoubtedly describe, if not the ‘same’, then at least related rituals.46 The descriptions are not identical, which may well be due to the knowledge or lack of knowledge of the authors, although it could also indicate that the entrance into blood-brotherhood could be ritualized in slightly different ways, to which we shall return shortly.
in the following, we shall deal briefly with the various elements and their possible symbolism, especially that of ganga undir jarðarmen.47 in the descriptions quoted above, we essentially see three rites: namely, the swearing of oaths, the mixing of blood, and the jardarmen rite. The swearing of oaths should probably be seen as a confirmation of the entry into blood-brotherhood directed both at the people involved (and probably also those attending) and at the gods (cf. the description in Gísla saga).48 We hear about oath-taking and the swearing of oaths in many sagas, although often without learning much about the accompanying rites (cf. Hellmuth 1975: 16–53). But the taking of oaths will, in a religious society, most often be part of religious rituals involving the gods, who should probably be seen as guarantors for the keeping of the oaths, and the oaths, therefore, will involve some numinous qualities. and the same is, of course, part of the symbolism connected to blood:49 blood has, as we saw in 46 it has been argued by, for instance, Rolf Heller (1976: 116–22) and others that the descriptions should be seen as literary ‘loans’. This cannot be ruled out, but similarities could just as well be due to knowledge of the same ritual procedure, of course with the variations that must always be expected.
47 The rite of ganga undir jarðarmen was apparently also used in other contexts than those of forming a blood-brotherhood (see, for instance, Pappenheim 1919: 70; Pappenheim 1924: 116). in Laxdæla saga ch. 18, it seems to serve as a Gottesurteil (judgement by the gods), and in Vatnsdæla saga ch. 33 as a kind of penitence caused by some shameful act (cf. also Brennu-Njáls saga ch. 119; see also schjødt 2008: 369–72). in later folklore there are examples of how parents let their sick children be drawn through a hole in the turf in order to cure them (olrik and Ellekilde 1926–51: i, 361), clearly implying that the child is reborn.
48 We should also remember here that in saxo’s description Óðinn is certainly involved, although it is not said directly that he was called upon during the ritual, neither as a witness nor as anything else.
49 The fact that blood is not mentioned in Fóstbrœðra saga cannot be taken as an indica-
32 – Passage Rituals
849
the previous chapter and in (è25), some sacred or numinous value, whether we are talking about sacrifices or the mixing of blood in connection with passage rituals. The blood of the participants is mixed, whether it runs into the palms of their hands and/or is mixed with the soil, most likely both, which is quite a powerful symbolism for the transfer of abilities. in Gísla saga, there is no mention of footprints as in Brot af Sigurðarkviðu and Gesta Danorum, but the blood is mixed with the soil underneath the turf, which could very well be in the footprints of the participants without the author explicitly stating it. The symbolism, however, seems clear: the blood, be it mixed directly in the palms or in the footprints, which have a metonymic relation also to the participants, creates a bond between them and this is of the same kind as that between blood relatives.
The jarðarmen rite seems to highlight the significance of the underworld with its numinous power, so that the blood in the footprints not only creates a bond between the participants, but it also creates a bond to the underworld realm. in this way, the participants symbolically go to the underworld, which is also the world of the dead (de Vries 1929a: 132), and in doing so they separate themselves from the ordinary space in which they usually live. This can easily be understood as part of the three-phase structure, suggested by van gennep, as a rite of separation: the participants ‘die’ out of the existence they had prior to the ritual. Their ‘return’ from the underworld should likewise be seen as a kind of rebirth, as has been suggested by many commentators, including Pappenheim, de Vries, and Hellmuth, and thus as a rite of incorporation. in between, the liminal part of the ritual, including the swearing of oaths that involve the gods and the mixing of the blood, has taken place, thus making the soil beneath the jarðarmen almost literally a liminal space.
With all the reservations we have to take into account when we attempt to reconstruct such rituals, we could perhaps imagine the following series of events: Turf is dug up from the soil to form a jarðarmen. This already indicates tion that it played no part in the ritual sequence. of course it cannot be ruled out that this could be the case, but it is much more likely, since the blood is mentioned in the three other passages, as well as in Brot af Sigurðarkviðu st. 17, that the author of the saga did not know this rite or perhaps did not find it interesting to relate. it is also somewhat suspicious that it is said that the passage under the jarðarmen in itself would signify the oaths, since, in the old Norse wording, the two elements seem more or less identical: to go under jarðarmen is their oath. This description, therefore, appears somewhat different from the others. Theoretically, we could propose that the authors of the three other passages added this incident with the blood, but it seems much more likely, as mentioned, that the author of Fóstbrœðra saga simply did not know about it.
850
Jens Peter Schjødt
that people other than the blood-brothers-to-be were part of the ritual, which makes good sense, also because it was probably important that there were witnesses (apart from the gods) to the formation of this new confraternity. Under the turf, a spear, perhaps a reference to Óðinn whose main attribute is the spear, is raised. Then, probably attended by some less spectacular rites, since they are not mentioned in any of the descriptions, the coming blood-brothers will descend into the symbolic underworld, maybe thinking of themselves as people about to die out of their ‘old existence’. Here, they will let their blood flow and it will be mixed, either directly on their palms or less directly via their footprints. We must suppose that the oaths were spoken, and other words, probably addressed to the gods, could perhaps also be involved. Having been changed in this dramatic way, the participants were then reborn to their ‘new existence’, which involved new relatives who were no less close than those of their former life. of course, we cannot know all these things for certain, but it seems safe to assume that in the pagan period the likelihood is that much more took place during these rituals than what we are explicitly told about.
Therefore, what we have in these descriptions is probably the most detailed information concerning any passage ritual in PCRN, except for the funeral ritual described by ibn fadlan. as was mentioned above, it seems somewhat surprising that the institution of blood-brotherhood was in this way remembered better and in more detail than initiation rituals into other, more spectacular, offices. Possible explanations could be that they grew out of the initiations into the military confraternities independent of the central authority of the kingdoms, which did not exist in iceland. Consequently, blood-feud remained more important in iceland, and for that reason knowledge of the ritual lived on to be recorded in literary texts emphasizing the need for vengeance. However, such a development is difficult to support from the sources.
Concluding Remarks
as we have seen, even if the sources are far from ideal, it is possible to reconstruct a few elements that seem to have been part of the passage rituals within PCRN, and this is also the case when it comes to the reconstruction of the structure and symbolism involved in the types of passage rituals dealt with above.
The sources do not allow us glimpses of other kinds of passage rituals or initiations, although we can be fairly sure that they did exist. for instance, the entrance into at least some of the various kinds of religious specializations (è29) presumably involved such rituals. it is, thus, not likely that people who specialized in runic magic should not have gone through some learning pro-
32 – Passage Rituals
851
cess, involving Óðinn and a number of rituals connected to him, or that people who were able to perform seiðr were not somehow exposed to those spirits that were to help them perform their art. The fact is, however, that these matters are not described in any written sources, and they cannot be detected, either, through archaeological material. it is only by means of comparative material that we can assume their existence, and therefore it is not possible for us to reconstruct such rituals in all their details.
33 – Death Ritual and Mortuary Behaviour
Neil Price
Death is a collective experience, one that includes the living left behind, who may also themselves be changed by the absence of the dead.
Burial customs are often analysed by archaeologists in terms of typologies of artefacts and grave form, assemblages of so-called ‘grave goods’ (the things accompanying the dead) and changing combinations of material culture. While all these and other, similar lines of enquiry can produce fruitful results, in fact they are all subsumed within a larger picture: burials represent behaviour, by no means only relating to the dead, and behind that behaviour lie entire worlds of ideas, beliefs, practices, customs, traditions, and social concepts of the normative and deviant. The later iron age of scandinavia and the northern cultures is especially fertile ground for the study of these worldviews, particularly as they encode loosely conceived packages of religion and spirituality, and offer a chance to attain a glimpse of these peoples’ attitudes to life as manifest in their approaches to death.
This chapter is intended to provide a critical review of this complex field, including a broad synthesis of the available material and its changing interpretation, with a forgiving eye to the conventions of the previous generations of research and a simultaneous caution concerning the conservative influence of received wisdom. The focus throughout will be on archaeological data, including the necessary reinterpretation of antiquarian finds and excavations from past decades. Written sources will also be brought to bear, but with a stress on Neil Price, Professor of archaeology, Uppsala 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. 853–896
BREPols
PUBlisHERs
10.1484/M.PCRN-EB.5.116960
854
Neil Price
contemporaneity and the context of their composition and transmission. in addition to a source-critical assessment of the relevant texts, attention will also be paid to the usefulness of the different perspectives that may be applied to them (see also è32).
The burial customs of later iron age scandinavia in particular exhibit a marked degree of social, regional, and chronological variation. strong emphasis will be placed on nuanced and multi-scalar interpretations, taking care to avoid assumptions of ‘religion’ in assessing the behaviour for which the graves provide evidence. funerary rituals will be explored as a component of the larger thought-world mentioned above but also embracing many other arenas of social action. furthermore, burial must not be seen as merely an act, a single event at a brief moment of time, but as part of a longer continuity both before and after.
The chapter will conclude with the need to assess the wider context of mortuary behaviour at every level, from its social situation to the literal landscapes of memory within which ‘cemeteries’ may prove to be illusorily discrete entities. Key themes must include agency, active interaction with the funerary environment, and, above all, the question of what it meant to be dead in the later iron age.
Dealing with Death: Research Approaches
to Late Iron Age Mortuary Behaviour
a comprehensive survey of scholarly paradigms in the interpretation of late iron age mortuary behaviour is of course impossible in a treatment of this scope, but we may profitably employ one key region — iceland — as a lens through which to view more general trends, with reference back to the broader picture in the scandinavian homelands.
This island in the North atlantic has long assumed a perhaps dispropor-tional prominence in the overall trajectory and reception of early scandinavian history, especially that of the Viking age, due to the unique preservation of old Norse narrative tradition in the form of the sagas and poetry ( Jónas Kristjánsson 1988; McTurk 2005; Clunies Ross 2010). This enormous corpus is of course a prominent element of the entire PCRN project and requires no further explanation here, but it is of importance to the study of funerary archaeology because of its influence on the entire perception of the Viking age past as it played out in the scandinavian nations during the eighteenth and nineteenth centuries. This process has been well studied (e.g. Wawn 1994, 2000; Roesdahl and sørensen 1996; Wilson 1997; gerhard and others 1997; svanberg 2003a), and in relation to mortuary behaviour it encouraged a romantic view of
33 – Death Ritual and Mortuary Behaviour
855
Viking heroes and noble shield maidens, visualized through finds from graves.
Coupled with this deeply rooted narrative heritage, especially prominent in the folklore and cultural life of rural communities, a strong tradition of what might be called ‘popular antiquarianism’ was prevalent throughout the North in the early modern period (adolf friðriksson 1994). in iceland especially, wherever remotely possible this resulted in a compulsively determined attempt to link the occupants of excavated graves to named figures in the Norse literature, or at least to the farms that they were thought to have inhabited.
This relationship between sagas and a kind of historical topography of burial was championed in the late nineteenth century by scholars such as Kristian Kålund (adolf friðriksson 2013: 28–35), and later incorporated with the developing school of archaeological typology and fledgling science by the society of antiquaries of iceland. The latter was strongly influenced by the work of swedes such as Hans Hildebrand and oscar Montelius, especially the latter who in 1895–97 produced the first refined chronology of the later iron age, largely derived from burial excavations. This work was taken up in iceland by researchers such as Valtýr guðmundsson. Towards the end of the 1800s and on into the early twentieth century, these new perspectives were given massive impetus by the discovery of the ship burials at Borre, oseberg, and gokstad, and the work of Norwegian archaeologists such as oluf Rygh, Nicolas Nicolaysen, Håkon shetelig, and a.W. Brøgger; in sweden, the exploration of the Vendel boat graves by Hjalmar stolpe had similar effects (see below for references to all these excavations). once more focusing on a cultural historical approach associating archaeological finds with the saga histories, burial excavations such as these brought an overwhelming focus on the high-status dead — on the chieftains and princes who were then thought to be emerging from the old Norse epics into empirically verifiable reality (something especially tangible at the gamla Uppsala cemeteries and other so-called storhögar in central sweden (lindqvist 1936). The complexity of these discoveries also began to open the way for systematic post-excavation programmes that would have far-reaching implications across scandinavia, not least in the excavations at gamla Uppsala, which was perceived as a comparable site in both mythologized legend and material culture to Norwegian Borre.
The new understanding of the sheer scale of evidence that was preserved in the scandinavian landscape led to a drive towards salvage and/or preservation, both somewhat innovative concepts in the early 1900s when what we would now call ‘heritage management’ was only in its infancy (adolf friðriksson 2013: 35–38). in iceland this was pioneered by Matthías Þórðarson, but the real breakthrough came with the work of his successor as state antiquary,
856
Neil Price
Kristján Eldjárn, whose efforts on behalf of the country’s heritage later took him to the presidency. Eldjárn conceived the idea of a new icelandic history, one based firmly on archaeological evidence rather than the received wisdom of the sagas. This marked a major break with previous traditions, and he soon put this into practice in the compilation of a massive catalogue of all Viking age graves then known from iceland. The resulting work — which in its various incarnations still forms the backbone of study in this field — was published in 1956, and was updated and expanded by adolf friðriksson in 2000 and 2017.
another major transformation in burial studies was felt throughout scandinavia with a beginning in the 1950s and accelerating impact from the 1970s onwards: the development of so-called ‘rescue archaeology’, in the wake of major infrastructure projects (see Bennett 1987: 5–14, discussing the gradually increasing contributions made by Mälar sites such as lovö, Helgö, and Birka, with numerous references; see also è6). far from focusing on spectacular monuments for the upper classes, the essentially random — or at least topographically driven — location of the excavations revealed hitherto unsuspected landscapes of death and brought about a new democratization in the study of late iron age mortuary behaviour. That this occurred by accident rather than design has not lessened its implications for our understanding of the period.
simultaneously with this new direction in funerary archaeology, saga specialists were beginning to seriously question the usefulness of the textual sources in the reconstruction of an actual Viking age reality (a discussion beyond the scope of this chapter, but see Clunies Ross 1994a, 1998a, 2010). one arguable consequence of this growing divide between archaeologists and textual scholars was a corresponding separation of interpretation (which was seen as the realm of narrative, and thus text) and data-gathering (viewed as the prime mission of archaeology). for many years, and all too often still continuing today, burials were therefore recorded, described, and published essentially as catalogues of the dead and their supposed possessions (in fact, the objects may have belonged to a variety of different people, and by their presence signalled different meanings accordingly). Thus we read of individuals interred with a number of artefacts and animals presented simply as an itemized inventory list.
furthermore, the artefacts themselves, the so-called ‘grave goods’, have been focused upon for their intrinsic interest, at the expense of other preserved traces that might instead point to what had actually happened in the course of the funeral, the presumed ceremonies, or rites. in particular, it is often overlooked that people, animals, and objects in the burials were in many instances only present in part, for example, in the form of deliberately broken artefacts, dismem-
33 – Death Ritual and Mortuary Behaviour
857
bered and incomplete bodies, and cremation deposits that clearly do not represent a whole corpse. When cremations are not found on the site of the pyre, where were the dead burned? in the case of animals, where were they killed?
Where only some body parts of animals or humans are present, what happened to the rest? all this necessarily implies that the ‘missing’ pieces must have been disposed of somewhere else, perhaps in several places. By extension, this would suggest that ‘funerals’ may have been very much more elaborate and extended processes, both in space and time, than has conventionally been assumed.
a related problem concerns the identity of the people in the graves. Even when skeletal remains are present and reasonably well preserved, osteological determinations of both sex and age are notoriously unreliable, and in any case say little of full-spectrum gender constructions. However, in the not uncommon absence of preserved bones, the sex and gender of the dead have often been — and still are — frequently inferred from objects alone, certain types and combinations of things being assumed to equate simply with biological sex (jewellery for a woman, weapons for a man, and so on). This misleading reliance on artefactual, osteological, and genetic signals has been analysed at length and with despairing accuracy by Back Danielsson (2007: 60–63 and ch. 2), who rightly observes that in most cases we are simply sexing metal and assigning gender to bones.
it is hard to overestimate the impact that these biases have had on our published record of late iron age burial, and thus our understanding of contemporary mortuary behaviour. The assumptions noted above are often undeclared in the literature, and without detailed re-examination of every individual grave in the museum storerooms of the North (in cases where the material even survives at all), we have little way of knowing to what degree the burial of, for example, ‘a woman in her mid-40s’ is in any way a genuine reflection of the actual grave or the person in it. in some instances from early excavations or antiquarian records, it is not always clear that artefactual sex determinations have been made in the absence of human remains, even when the dead are firmly identified as male or female. it is clear that no ready solution presents itself.
in the text that follows here, the sex determinations of the original excavators are employed for want of alternatives, and where any ambiguity is present and detectible in the grave itself, this is brought out in discussion and in full awareness of its uncertainties.
a representative overview of current Viking age funerary archaeology is impossible, due to the encouraging variety of perspectives and positions.
However, a fewer broader trends can be noted, with a caveat that this is merely the tip of a vast iceberg.
858
Neil Price
in recent years there have been tendencies to approach graves very much at the level of the individual, not only as a unique window onto a specific life but also as a personal lens through which to view the time as a whole (Harrison and svensson 2007; Price 2010). This has been employed with particular success in communicating mortuary behaviour to the public, linking attitudes to life and death, as at aarhus where ‘sleeping’ Vikings introduced exhibition visitors to early medieval aros (skov and Varberg 2011), and at Birka in the form of a five-year-old girl’s grave and subsequent reconstructions (Hedenstierna-Jonson 2012a).
This new concern for specificity has also led some scholars (e.g., Back Danielsson 2007) to a move away from mortuary identities based on biological sex in any sense, and a search for a more intensely somatic and mutable past; others seek new levels of nuance in more traditional, gender-based readings of graves (e.g., Hayeur smith 2004, among many others). another trend, linked to wider theoretical developments in archaeology as whole, emphasizes the materiality of graves and their contents, the assemblage of identity through things, and the ways in which the human and material interact and define each other (e.g., Þóra Pétursdóttir 2009). another approach involves the search for plural-istic narratives encoded in the rituals that can be reconstructed from funerary archaeology, conceptualized not in relation to formal structures such as sagas, but as personal responses to death and its meaning for the living (e.g., Price 2008a, 2008b, 2010, 2012). in particular, careful recording and stratigraphic analysis have been championed for their potential in retrieving ritual sequence and character, the process of the funeral rather than merely its result (gansum 2004; Bäck 2011; Ulriksen 2011; Price 2014).
Using this compacted research history as a platform, we can review in turn the specific textual and archaeological evidence for Viking age funerary practices, and the thought-world form which they emerged.
Eyewitnesses and Hindsight
There is no doubt whatsoever that archaeology provides our primary evidence for the burial practices of the later iron age, in all their immense variety, and this is reflected in the emphasis of this essay. However, it would be unwise to ignore the fascinating but deeply problematic information that can be gleaned from close study of textual sources, to which some short notes can be devoted here. We possess very few direct descriptions of late iron age pre-Christian burial customs, but as we have seen this has not deterred scholars from apply-ing this minimal corpus far more widely than is probably warranted. That said,
33 – Death Ritual and Mortuary Behaviour
859
they are also a mine of useful data, which with appropriate source critique can be brought to bear on archaeological problems. That the excavated material should not only corroborate but also to an extent sharply contradict the textual sources should not surprise us, but of key importance is the fact that the archaeology reveals mortuary practices that have left no documentary trace at all.
The relevant texts fall into two essential categories: a small, primary group of eyewitness accounts, mostly from arab and Byzantine writers; and a larger group of secondary descriptions in icelandic sagas and a handful of old Norse poems, written down centuries after the late iron age and of a value that has long been debated across a broad spectrum of perspectives (see also è3 and è32).
The arab writer ibn fadlan’s description of a Rus ship burial is very long and will not be reproduced here (numerous translations have been published in whole or part; see lunde and stone 2012 for the most recent complete edition, and Montgomery 2000 for a scholarly text with references to the original; see also è32 for a brief summary). Unsurprisingly, this account has been widely discussed by Viking specialists, and also by archaeologists working more generally with burial studies (e.g., arne 1941; Ritter 1942; Kowalska 1973; Warmind 1995; Parker Pearson 1999; Taylor 2002; schjødt 2007b). Neil Price has dealt extensively with ibn fadlan’s narrative (Price 2002, 2008b, 2010, 2012), and has focused upon: ibn fadlan’s description of the ten-day sequence of the rituals; the employment of a temporary grave for the corpse while the ship is made ready; the manufacture of objects intended for burial; the movement, gestures, and sounds involved; the use of food and especially alcohol; the killing of animals and humans; the frequent sexual performance; the detail of the graves’ contents and their order and placement; and the actions of the various individuals present. some aspects of ibn fadlan’s account will be taken up below, but it is clearly an essential source for anyone interested in Viking age mortuary behaviour.
other arab writers of interest include ibn Rustah and ibn Miskaweih, who describe how women might be buried alive in the chamber graves of their male partners, and who also record the deposition of swords in the graves of warriors (Birkeland 1954; Montgomery 2008).
The Byzantine sources also preserve several items of interest, particularly from accounts of a campaign against the Rus in the early 970s, a force with very strong cultural links to scandinavia. one episode, mentioned in John skylitizes’s Synopsis of Byzantine History 15.12, occurs during a lull in the fighting when the Rus army is surrounded by the imperial troops after a battle. Here it says that the Rus remained awake all night along, mourning for those who had fallen, and that it sounded more like the roaring and bellowing of wild
860
Neil Price
beasts than the grief and lamentation of humankind. although given but a passing mention, it is interesting how the Rus clearly mourn their dead at night.
another description of the same occasion comes from the greek historian leo Diaconus, who accompanied the Byzantine emperor as secretary. in the eighth and ninth books of his Historia, he notes first how the Rus emerged from their besieged fortification at night to retrieve their dead from the previous day’s combat. He describes how the Rus went among the slain in the light of the full moon, building great pyres of logs outside the wall, on which they cremated the dead. Male and female prisoners were brought out and sacrificed, along with young animals. Cockerels were killed and thrown into the river, and offerings were made of what appears to have been alcohol (Ellis Davidson 1972: 25).
Turning to the retrospective sources (see also above è32), the old Norse prose and poetic corpus contains several mentions of what are claimed to be Viking age burials, almost all incidental details in wider narratives. Very few of these instances make direct reference to any beliefs that might lie behind the practices and situations described, though it is possible to gain some approxima-tion of ideas attaching to the dead and their condition. Before briefly considering a representative sample of these, we can turn to the single deliberate account of burial practices that also contains some explicit statements about their meaning.
Ynglinga saga ch. 8 presents the so-called ‘laws of Óðinn’ in which he decrees the proper treatment of the dead. it is worth quoting in full: Óðinn setti lǫg í landi sínu, þau er gengit hǫfðu fyrr með Ásum. svá setti hann, at alla dauða menn skyldi brenna ok bera á bál með þeim eign þeirra. sagði hann svá, at með þvílíkum auðœfum skyldi hverr koma til Valhallar sem hann hafði á bál, þess skyldi hann ok njóta, er hann sjálfr hafði í jǫrð grafit. En ǫskuna skyldi bera út á sjá eða grafa niðr í jǫrð, en eptir gǫfga menn skyldi haug gera til minningar, en eptir alla þá menn, er nǫkkut manns mót var at, skyldi reisa bautasteina, ok helzk sjá siðr lengi síðan. (p. 20)
(in his country Óðinn instituted such laws as had been in force among the Æsir before. Thus he ordered that all the dead were to be burned on a pyre together with their possessions, saying that everyone would arrive in Valholl with such wealth as he had with him on his pyre and that he would also enjoy the use of what he himself had hidden in the ground. His ashes were to be carried out to sea or buried in the ground. for notable men burial mounds were to be thrown up as memorials.
But for all men who had shown great manly qualities memorial stones were to be erected; and this custom continued for a long time thereafter.) (pp. 11–12) Written two centuries after pre-Christian mortuary behaviour was the norm, in isolation we have little way of evaluating the degree to which the ideo logical
33 – Death Ritual and Mortuary Behaviour
861
filters of his own time shaped snorri’s presentation of these rites. However, from the similarities with excavated examples, it is clear that to some degree the descriptions are based either on some kind of actual collective memory, and/
or practical knowledge of pre-Christian burials from empirical observation of their remains. With these caveats in mind, it is worth drawing out some key points in Óðinn’s alleged commands:
– all the dead are to be burned, irrespective of status; there is no mention of inhumations
– the things burned with them are their possessions, which would follow the deceased to the next world
– all the dead are apparently bound for Valhǫll (or is it just those whom this ancestral Óðinn governs?)
– things buried unaccompanied by bodies are also for use after death, and can be selected for that purpose by the living, making arrangements for their own afterlives
– burial mounds are only for special, ‘notable’ men (it is hard to know how androcentric this statement actually is meant)
– standing stones are for the commemoration of anyone of note
– at least some of the cremated dead are disposed of in water, that is, in a manner invisible to archaeology; it may be significant that this deposition of ash in the sea is the first method of disposal that is mentioned — and thus the most common?
after more than a century of excavations, there can remain no doubt whatsoever that we cannot speak of a standard orthodoxy of burial practice common to the whole Norse world: snorri’s ‘laws of Óðinn’ are an illusion, even for the rather vague ‘country’ (a kind of fictive sweden) to which they allegedly applied. This does not mean that every part of his description is inaccurate, but instead we should examine it in specific rather than generalized contexts.
We will encounter several of these qualities in the excavated material, and the possible light that snorri’s account may shed on them should not be discounted (see lindqvist 1920 for an extensive early echo of this debate).
in total the saga corpus contains some 150 references to burial, which is intrinsically interesting in that there is no self-evident reason why this should be discussed at all. it is clear that graves were important in the late Norse mind, and also that their location was an integrated part of their meaning (adolf
862
Neil Price
friðriksson 2013). Earthen mounds are frequently mentioned, and the sagas also contain references to a significant number of what are clearly chamber graves, inhumations within wooden rooms, built either above or below ground and covered by a barrow. The degree to which the mounds’ incumbents were still thought to ‘reside’ in their graves, and thus remain members of their communities, is arguable, though their metaphorical presence seems assured. The old Norse prose sources contain many stories of the living dead in the sense of the physically reanimated corpse, but while the majority of these tales concern evil beings, there are also a significant number that merely relate how the dead live on in their graves (cf. sävborg 2011). a prominent example is that of gunnarr of Hliðarendi from Njáls saga, who in Chapter 78 is seen happily singing in his mound one night, seated and surrounded by four ‘lights’ that may be candles (è34). similarly, the dead warriors of the unsettling poem known as The Waking of Angantyr seem to sleep uneasily in their burials, ‘down among the tree-roots’ (the poem is eddic in style but found in a later story, Hervarar saga ok Heiðreks konungs; Terry 1990: 248–53; larrington 2014: 268–73).
The most vivid account, and the most archaeologically useful, comes from Grettis saga Ásmundarsonar ch. 18. The anti-hero grettir has decided to rob a large burial mound on a headland, over which he has seen flames hovering at night, and assisted by the local farmer he begins to break into the mound from the top. He works hard until he reaches the ‘rafters’ ( viðir), which he then breaks through. lowering a rope, he prepares to enter the barrow: gekk grettir þá í hauginn; var þar myrkt ok þeygi þefgott. leitast hann nú fyrir, hversu háttað var. Hann fann hestbein, ok síðan drap hann sér við stólbrúði ok fann, at þar sat maðr á stóli. Þar var fé mikit í gulli ok silfri borit saman ok einn kistill settr undir fœtr honum, fullr af silfri. grettir tók þetta fé allt ok bar til festar; ok er hann gekk útar eptir hauginum, var gripit til hans fast. lét hann þá laust féit, en rézk í mót þeim, ok tókusk þeir þá til heldr úþyrmiliga. gekk nú upp allt þat, er fyrir varð; sótti haugbúinn með kappi. grettir fór undan lengi, ok þar kemr, at hann sèr, at eigi man duga at hlífask við. sparir nú hvárrgi annan; fœrask þeir þangat, sem hestbeinin váru; kippðusk þeir þar um lengi […]. (pp. 57–58) (Then grettir went inside the mound. it was dark and smelled unpleasant. He explored the mound to see how it was laid out. He found some horse bones, then he rubbed against the carved back of a chair and could tell there was a man sitting in it. a huge amount of gold and silver had been piled up there and the man’s feet were resting on a chest full of silver. grettir took all the treasure and carried it over the rope. and when he was walking back inside the mound, something grabbed him tight. He dropped the treasure and fought back, and the two of then grappled violently, knocking everything over that was in their way. The mound-dweller went
33 – Death Ritual and Mortuary Behaviour
863
for him ferociously, and grettir backed off for a long time, until he realised that he would need all his strength. They both fought with all their might, and struggled towards where the horse bones were. They grappled for a long while […]). (p. 75) The roof-construction of the chamber is exactly paralleled by excavated examples (see below), as are many of the contents and their disposition, including the seated incumbent, boxes of precious objects, and the presence of a horse slightly separate from the main chamber.
Barrows such as this also feature in some of the eddic poems as the site of supernatural encounters, and in skaldic verses as places where it was possible to gain a strange kind of heightened receptiveness to the influence of otherworldly powers. it is clear that after the burial, these landmarks — sometimes named, in the manner of Anundshög, ‘anund’s Mound’ — played a significant part in the cognitive landscape of the community.
Moving outside the strictly scandinavian corpus of texts, though to a story nonetheless set there, we find that one of the greatest northern epics of all, Beowulf, is actually structured around funerals (owen-Crocker 2000; fulk and others 2008). The poem opens with the burial rites of a king, taking the unique form of his deposition in a ship that is then pushed out to sea, a scene that has inspired countless popular culture imitations, despite the rarity of the original. This beginning is matched by the epic’s end, with the mound burial of the titular hero. This is described in detail, highlighting the deposition of precious metals, the formal mourning of the dead including a ride past by leading retainers, the considerable length of the rituals, and the importance of visibility in the finished barrow. although definitely written down in the Viking age, its scribal ‘author’ was an English Christian relating a tale from the scandinavian Migration Period, so — like the sagas — its usefulness as a genuine source is problematic.
for the mythological sources and their descriptions of funeral rites, we refer to (è32).
Variation and Consistency
Perhaps the central element of Viking age scandinavian funerary ritual was its individual character, most evident in the artefactual assemblage but also in the rites or ceremonies themselves. The selection, combination, particular type, quality, quantity, and exact positioning of this material are all factors in the variation within Viking age mortuary ritual. all across this spectrum of behaviour, it is vital to note the detail that can give us information about the
864
Neil Price
sequence of events, the time they may have taken to perform, and the spatial arena of ritual that in some cases must have extended very considerably beyond the grave itself.
in addition to the variation between individual graves, it is important to understand that at a larger scale this also forms patterns of regional, even local, expression. This in turn must imply at least a degree of variation in the meaning behind these practices, and thus the structures of belief relating to the treatment of the dead. No comprehensive survey is possible in an essay of this kind, but some of the broader trends can be noted.
one of the first, and still most important, studies of specific burial practices right across scandinavia was undertaken by Johan Callmer (1991b, 1992), who over many years mapped the excavated funerary assemblages at a cemetery-level of resolution. in so doing he was the first to empirically demonstrate that variation was present at the level of individual communities, villages, and even extended farmsteads. from one settlement to another people handled the dead in broadly consistent ways — essentially through cremation or occasionally inhumation — but differed in the details of grave construction and elabora-tion, the placement of the body, and the selection and deposition of objects that accompanied the deceased. a similar survey has been undertaken more recently for finnish mortuary behaviour by anna Wessman (2010).
it has been suggested that this diversity is a signal not of varying treatment of the dead within a single society, but it is instead evidence for the illusory nature of the ‘Viking age’ itself: that the highly regional burial traditions are indicators of distinctive ethnic, social, or political groupings that make a mock-ery of the notion of a pan-scandinavian culture (svanberg 2003a, working from his superbly detailed analyses; svanberg 2003b, on discrete cultures of ritual in southern sweden). The problem with this interpretation is that it ignores the very real, general similarities of material culture within the region (not to mention language and settlement pattern) and focuses only on variations that are nonetheless practised within a broader, consistent framework. That villages or even larger communities promote their own identities does not mean that they have no part of larger ones. The culture of the Viking age scandinavians is as evident in their burials as in other aspects of their society.
Particular forms of burial, treated in more detail below, also have regional trends. frans-arne stylegar’s work on chamber grave traditions (2005, 2014) has demonstrated a clear concentration in the western fjords of Norway, with very few found around the oslofjord and the east. a related form, known in Norwegian as hellekister or ‘flagstone coffins’ — effectively a stone chamber grave though with more elongated outline — is located almost solely in the
33 – Death Ritual and Mortuary Behaviour
865
figure 33.1. Burial ground consisting of mounds at Kånna in finnveden (south-western småland) from the late iron age. These mounds provide a good example of localized burial customs, which varied between different regions. Photo: anders andrén.
west. in some hybrid cases, wooden chambers are built inside the hellekister, and this type is found only along the south-west coast with a single, northern outlier at Veiem in Nord-Trøndelag. This pattern holds broadly true from the Roman iron age and on into the Viking age, with gradual changes in the exact structure of the graves through time. in sweden, chamber graves are instead found focused on the Mälar Valley and Uppland province, with a special focus on urban contexts such as Birka (see below). in all cases, they are likely to be the burials of individuals who enjoyed a markedly higher status than the norm.
There are also numerous examples of localized practice, of which only a few can be referenced here to give a flavour of the ritual range. in eastern central sweden and Åland, for instance, we find a unique emphasis on the burial inclusion of so-called Thor’s hammer rings. Usually placed on top of cremation urns, this ritual is found only in the late 900s and into the eleventh century, which perhaps suggests that it developed in direct response to the rise of Christianity in these areas at that time — a demonstrative paganism in opposition to the new faith (andersson 2005b). in some areas of central sweden there is a clear emphasis on the inclusion of bear pelts in graves (Petré 1980), and so on.
some variations are extreme and clearly depend at least in part on the vagar-ies of preservation and excavation location. The central Danish island of fyn,
866
Neil Price
figure 33.2. a cremation grave
(grave a 65) with a Þórr’s
hammer ring placed on top of
the urn, from the burial ground
in söderby on lovö, Uppland.
after Petré 2011: 240.
for example, has only two confirmed burials of Viking age date, which cannot possibly reflect a complete reality but may nonetheless indicate a differing treatment of the dead there, perhaps tending more towards burial without trace (see below).
We also find special rituals in island communities, and in general the funerary rites of places such as gotland, Öland, Bornholm, and Åland are unlike those of their respective closest mainlands, which differ in turn from the surrounding areas (e.g., Thunmark-Nylén 1995–2006; Beskow sjöberg and others
33 – Death Ritual and Mortuary Behaviour
867
figure 33.3. Burial ground at Trullhalsar in anga on gotland, dated to the late iron age.
The burial ground, which is one of the best preserved on gotland, consists primarily of cairns, which are typical of gotlandic graves. Photo: anders andrén.
1987–2001). on gotland in particular there are tendencies not only towards very large cemeteries with a long continuity over time, with a particular emphasis on inhumation (such as ire and Barshalder; see Rundkvist 2003), but also for clusters of burials with unusual mortuary behaviour, including prone inhumation, decapitation, and complex ritual trauma (such as lilla Bjärge, fröjel and Kopparsvik — see Wickman-Nydolf 1999; Carlsson 1999; arcini 2010; funegård Viberg 2012). Here too there are localized rituals, as in the use of colour symbolism in burials of young girls interred with white beads made of exotic cowrie shells, that may suggest new connections between material culture and visible status (Thedéen 2010).
on Öland, for example, fossils such as ammonites were sometimes deposited with the deceased (Beskow sjöberg and others 1987–2001). on the Åland islands between sweden and finland, the ashes of the dead were buried in pottery vessels, on the top of which was placed a miniature animal paw made of clay. The paws, which were not present on the funeral pyre, have been identified
868
Neil Price
as characteristic of either bears or beavers. This rite is found only on Åland, and in specific clusters of graves on the Volga and Kljaz’ma rivers in Russia; from the accompanying grave goods, these burials have been convincingly interpreted as those of travelling Ålanders (Callmer 1994).
Recognizably scandinavian burial traditions are also found across the Viking world, again with local traditions in evidence. in the North atlantic colonies, cremation is extremely rare, and indeed very few graves of any kind from the period are known from the faroes (see papers in arge and Mortensen 2005). on iceland some 350 burials from the Viking age are known (Kristján Eldjárn and adolf friðriksson 2017), mostly located either singly or in small clusters in relation to farmsteads and along lines of communication (adolf friðriksson 2013). a number of boat burials have been found, especially in the north of the country (Roberts 2009), and the funerary assemblages in general contain fewer objects than on the scandinavian mainland (though this should not necessarily be taken to imply major differences in status; Þóra Pétursdóttir 2009). No scandinavian burials have so far been found in North america.
around the British isles, burials with objects of Norse type are concentrated in the north and west, with major concentrations around Dublin, Man, and orkney. The burials are almost entirely inhumations, including several boats, and are predominantly found in the coastal areas where the Norse settled. furnished female burials are more common in scotland than elsewhere in Britain (the material from the northern and western isles, and mainland scotland, is summarized in graham-Campbell and Batey 1998, where more recent finds include a remarkable child burial at Balnakeil; see also Batey and Paterson 2013). Mound burial, including the use of boats, is frequent on the isle of Man (Bersu and Wilson 1966; D. Wilson 2008), and the island also boasts several runic memorial stones. it lies at the centre of the irish sea cultural sphere, which shows considerable consistency in its burial rites of scandinavian and local character (griffiths 2010: ch. 5).
More than a hundred furnished inhumations are known from ireland, with a mixture of insular and scandinavian graves-goods (Harrison and Ó’floinn 2014). Much of the material was collected by antiquarians after being disturbed during construction work, especially around Dublin, and the analysis of the corpus has consequently been beset by documentation problems. However, a clear focus on Dublin is apparent, with discrete cemeteries visible around the town and the settlements that preceded it. The large numbers of recorded weapons testify to the nature of the Viking forces and their early impact on ireland.
in England there are almost no burials resembling the mounds common in scandinavia (see Halsall 2000). The main exception is the cemetery at Heath
33 – Death Ritual and Mortuary Behaviour
869
Wood with its fifty-nine barrows, which include animal sacrifices and a number of empty ‘graves’ of indeterminate meaning, with a possible link to the nearby winter camp of a Viking army at Repton, itself with numerous burials (Biddle and Kjølbye-Biddle 1992, 2001; Richards 2004b). Burials of arguably Norse character are otherwise concentrated in the north-west and in Cumbria (e.g., Paterson and others 2014), with very few in the east or south of England (Richards 2008). in the later Viking age, however, new classes of stone memorial are introduced (Bailey 1980), including distinctive grave covers known as
‘hogbacks’ (lang 1984), and even as far south as london there are elaborate runestones commemorating the new anglo-scandinavian elites.
in Continental Europe, very few graves can be unequivocally interpreted as of Norse origin. in the territory of the frankish Empire only a handful are known, principally from Normandy and Brittany, including the great ship burial of the Île de groix, discussed further below (Price 1989, 2013a; Renaud 2000).
in the eastern areas of Viking expansion, Norse funerary rituals are found amalgamated with slavic, Khazar, and other ethnic practices. alongside individual graves and small groups of burials, the cemeteries of the Dvina, Dniepr, Volkhov, and Volga river systems are both vast and complex — in some cases much larger than anything found in scandinavia — and in this they reflect the ethnic melting pot of their formation. The degree to which scandinavian influence is detectible there has been debated for centuries and lies at the heart of Viking age research in the East, but a detailed discussion is beyond the scope of this chapter (for recent attempts, see Duczko 2004; androshchuk 2008, 2013).
The Great Absence: Death without Trace
Before discussing specific rites for the burial of the dead, it is important to mention an aspect of Viking age mortuary behaviour that is often overlooked: quite simply, it is clear that not everyone was afforded a grave at all. Estimates of the proportion of the populace without a detectable burial place are unreliable, and essentially a matter of guesswork in attempting to correlate settlement scale (and estimates of how many individuals lived in an ‘average’ building) with the quantity of visible funerary monuments. However, it is not impossible that more than half of the population did not receive a burial that we can now detect.
it is perhaps reasonable to assume that these ‘missing’ dead were marked by low status, either the very poor or slaves, but we cannot be sure. We have no identifiable evidence for the burials of slaves in their own right, as opposed
870
Neil Price
to their presumed presence as sacrificial offerings in a few cases treated below.
Whether these people were cremated and their ashes then scattered or disposed of in water, or whether they were excarnated, is impossible to say. it is worth noting that these archaeologically invisible burial forms are mentioned not only in Ynglinga saga ch. 8 as we have seen, but also in first-hand accounts left by arab writers such as ibn fadlan, who described in the tenth century how dead slaves were simply abandoned, at least while on the move (Montgomery 2000; lunde and stone 2012: 49).
Children are also under-represented in the burial record, which may reflect a number of factors. We know little of how the child-adult transition was regarded at this time, and accordingly whether dead children were seen as ‘worthy’ of formal burial (Mejsholm 2009); the fact that we have child burials at all suggests however that the same criteria of familial and personal status may have been applied. The practice of child exposure and abandonment may also account for a large number of the children missing from the archaeological burial record (Wicker 1998).
Burning and Burying
There is no exact figure on the number of extant burials from the Viking age in scandinavia. on the imprecise basis of a professional lifetime working with the archaeological material, and following a personal canvas of Nordic colleagues who have done the same, a guess of perhaps half a million graves seems reasonable though open to question. By the same informal measure, one can reach a figure for excavated examples in the low tens of thousands.
They fall into some basic, broad patterns, but essentially the Vikings either burned their dead or interred their bodies. in sweden only cremation occurs, with occasional inhumations and chamber burials at exceptional sites such as Birka. in Norway and Denmark a mixture of cremation and inhumation was practised. We do not know why one ritual was chosen above another, though there is some, very tentative speculation that part of the intention with cremation was to prevent the dead person’s resurrection in some malevolent form. as with so much of Viking mortuary behaviour, this does not entirely square with the alternative in the form of inhumations, but does make sense when compared with later graves of the Christian period, when we know that the body was intended to return after the day of Judgement.
although most graves contain a single individual, on occasion two, three, four, or more bodies are found together, either interred or burned at the same time, or in successive depositions (selected examples are considered below).
33 – Death Ritual and Mortuary Behaviour
871
figure 33.4. Excavation of a grave at gnista in the parish of Danmark in Uppland.
The grave (no. 42) consisted of a burnt layer with cremated bones placed in an urn.
Photo: fredrik Thölin saU/Upplandsmuseet, Uppsala.
for those who received a visible burial, the most common means of disposing of the dead was through cremation, followed by the interment of the ashes either in unmarked graves or under mounds. Remains of pyres have been found on several sites, especially in swedish Uppland, sometimes with extensive survival of carbonized material. They most often seem to consist of cross-timbered wooden platforms, with plenty of space for air to circulate between the logs and provide a draught to keep the fire going. The dead — both human and animal — were either laid on top of the pyre, or else inside as it was constructed around them. in rare instances, the dead were burned directly on the ground surface.
Even at this stage of the rituals, it must be noted that a burning pyre was a dynamic, chaotic environment. The topography of the site, the weather, the fuel in the form of timber and kindling, the placement and condition of the body all played a role, and the result could only have been an emotional experience for those involved. fragments of shattered flint found in pyre remains hint at other dimensions too, as this stone explodes in dramatic showers of sparks when burnt at high temperatures, and it may have been added to cremations for just this purpose. fire and smoke can be seen over long distances, especially if the pyres are on high ground. We have already observed the possibility that Viking funerals may sometimes have been conducted at night, further heightening the visual impact of flame and firelight.
The variable predictability of cremation pyres requires considerable knowledge and experience to manage, and it is likely that special individuals had this task in the Viking age. in rural grave-fields, it is evident that many of the pyre remains found by archaeologists were from very badly built structures, perhaps
872
Neil Price
resulting from funerals conducted by family members performing these ministrations for the first time. Near larger settlements, there is clear evidence that the pyres are much better constructed, as we might expect in localities where funerals were a more common event. When the excavated remains indicate that the pyre has been consistently turned over and maintained at a constant temperature, this is a virtually certain indication of the presence of specialists.
after soft tissues, clothing, and so on have been burned away in the fire, it is not uncommon for skeletal material to be left charred but quite substantially intact after a cremation, more than might be assumed to be the case. in most instances the bones of the humans, and sometimes the animals, have been retrieved from the ashes, sorted, cleaned, and sometimes crushed, before being laid back on the remains of the pyre — either directly or in a container such as a ceramic vessel, a box, or a bag. Burial pits could also be dug down through the pyre to accommodate the ashes; in one unique case, at Klinta on Öland, a hex-agonal ceramic lid had been placed over the pit, under the remains of the pyre (Petersson 1958, 1964). another relatively common alternative saw the ashes removed for separate burial elsewhere. in cases where several people were cremated together, it is not unknown for the remains of the humans and animals all to be sorted separately and then buried in their own individual graves some considerable distance from the pyre, as again at Klinta.
This processing of the dead mentioned above, sorting the charred bones and ashes, is significant because in many — even most — cases it is clear that only a portion of the remains was subsequently placed in the ‘finished’ grave.
By way of demonstration, a modern, professionally managed machine-operated cremation reduces a male human body to some 7–8 litres of ashes and bone material, and slightly less for a woman; in late iron age cremation graves, it is uncommon to find more than about a litre, and even the largest ceramic cremation vessels can hold no more than 1.5 litres. We do not know what was done with the rest, but many interesting possibilities exist. Perhaps parts of the bodies were distributed to the family, friends, or comrades of the deceased; perhaps the limited quantities of ashes that we find in graves collectively add up to whole individuals, divided between what we have archaeologically recorded as distinct, separate burials; perhaps some of the ashes were strewn on the fields, in or around houses and buildings, in water. We will probably never find out, but it is clear that our present understanding of what constituted a ‘grave’ might need revision.
Throughout the Viking world it was very common for objects to be burned together with the dead, laid out around them on the pyre or placed about their persons. in the case of burials constructed over the pyre, even when the ashes
33 – Death Ritual and Mortuary Behaviour
873
figure 33.5. Woman buried in a carriage at a small burial ground by the ringfort fyrkat in northern Jylland from the late tenth century.
after Roesdahl 1977: 98, for a reconstruction, è figure 22.1.
have been treated and manipulated as above, the objects seem to remain where they fell vertically through the pyre as it burnt out. in some cases objects were deliberately broken before the pyre was lit, perhaps to mark their ‘death’ alongside that of their owner.
sometimes the ashes are also overlain by unburnt items, placed there during the construction of the grave. This combination of material from the pyre and from outside it is not uncommon, and provides further evidence for the complexity of the procedures involved, with a ritual meaning content that remains almost impossible to access.
inhumation was rarer but occurred across scandinavia. it has been argued that in the later Viking age, some of these burials represented transitional Christian graves, but this is debated (gräslund 1980, 2008b). Bodies were generally laid in rectangular grave cuts, either directly on the ground, on textiles or mats of bark (the latter especially in northern Norway), in shrouds, or in coffins of various kinds including the detachable cargo bodies of wagons (especially for women; Hägg 2009). The dead could even be interred in small boxes, with bodies folded and crammed into them — perhaps the defleshed remains of those who had died far from home, and brought back to the family farm.
Many different body postures have been found, though the dead are most often laid out either supine or slightly curled over on their sides, as if sleep-
874
Neil Price
figure 33.6. one of the richly furnished inhumation
graves at Himlingøje in stevns on sjælland, dated to
the third century ce (Nationalmuseet no. C24137).
Photo: Roberto fortuna and Kira Ursem, National-
museet, Copenhagen.
ing, with legs straight or flexed. in some
graves the remains of blankets, pillows,
and other bedding have been found under
and around the bodies, reinforcing this
suggestion.
some remarkable combinations are
found, such as a person buried on their
back with one leg bent at the knee and
turned. in a few unusual instances the dead
are buried prone or in a variety of unnatu-
ral postures that necessitate actual damage
to the body, for example, by the removal of
limbs. Whether this relates to some kind of
punishment or legal censure is hard to say,
but the large stones placed on top of some
of these bodies imply a fear that they might
somehow leave the grave and presumably
cause harm to the living. This raises obvi-
ous questions of deviation in relation to
the normative in Viking age burial, and of
course the problems in defining those con-
cepts in context (Murphy 2008; Reynolds
2009). a special study of such burials
in scandinavia and the Viking colonies
has been made by leszek gardeła (2011,
2012, 2013), reviewing a wide range of rit-
ual trauma, with the interesting conclusion
that such practices were very much more
widespread than previously believed, and
essentially form one expression of mortu-
ary behaviour among many others.
There are again regional variations in
all this, as in Norway where people were
33 – Death Ritual and Mortuary Behaviour
875
sometimes buried in large ‘coffin boxes’, taking the form of oversized graves lined and floored with wooden planking in a construction that had been built inside the cut rather than being lowered into it (shetelig 1912). often lined with mats of birch bark covered with textiles, as we have seen with the hellekister above, these same features can also be built of stone (stylegar 2014). in an increasing number of burials, especially in iceland, evidence is now being found for post-built structures erected over inhumation graves. Resembling small buildings, some of them also appear one-sided and even with posts inserted at an angle, perhaps making some kind of shelter-like affair above the interment (Þóra Pétursdóttir 2009; Roberts 2009).
inhumation burials normally exhibit the same or even greater range of grave goods as the cremations, though the apparent profusion is perhaps a factor of preservation. like the cremations, inhumations were also accompanied by animal and occasionally human offerings, along with considerable quantities of foodstuffs and, to judge by the containers, drink as well. among the artefacts deposited with the dead, the most commonly encountered include items of personal dress and ornament such as: jewellery; weapons; implements for textile production and food preparation; smithying tools; agricultural implements; household utensils, containers, and fixtures of various kinds; horse equipment; furniture including beds, chairs, and stools; textiles of varying quality and quantity; food and drink, amongst many other kinds of objects.
The Visible Grave
Beyond the level of cremation or inhumation, a number of clear categories of grave form can be discerned. The most ubiquitous is that of burials under mounds. Burials are found within stone settings, and also in the form of chamber graves, the iconic ship and boat burials, and occasionally as mass graves.
Within these basic structures, however, the variety in the detail of mortuary behaviour is almost such as to make every grave slightly (and sometimes dramatically) different.
Mound-Dwellers and Settings of Stone
Burial mounds could be of widely varying shapes and sizes, ranging from low humps in the ground to monumental barrows up to 10 m high or more.
Circular forms predominate, but oval, rectangular, and triangular mounds are also known. in some instances the mounds are augmented by what appear to have been posts set up in them, for unknown reasons, or by small pits dug
876
Neil Price
figure 33.7. The ship-setting ‘ales stenar’ at Kåseberga on the south coast of skåne is the largest preserved ship-setting, measuring 67 m long. it was probably constructed in the sixth century.
Photo: anders andrén.
into the sides, again of indeterminate purpose but presumably relating to the extended rituals of the burial, discussed further below. The relative numbers of burials in the same place can vary greatly, partly in natural relation to the size of settlements — an obvious difference between, for example, a town and a single farmstead — but also sometimes grouped in clusters that must have served more than one community. Mounds occur singly or in small groups, all the way up to cemeteries of thousands.
in general burials seem to have been unmarked in the sense of personally recording their occupants, but ibn fadlan’s account, mentioned above, describes how a mound was topped by a wooden pole, on which was cut (presumably in runes) the dead person’s name and that of his lord. leaving little archaeological trace (with some exceptions, as at Repton in Derbyshire; see Biddle and Kjølbye-Biddle 1992, 2001), this form of commemoration might have been more common than we suppose, and may also explain some of the postholes found in barrows.
in other ways the marking of graves was elaborate and widespread, and usually achieved with stones. These range from individual bautastenar, standing stones erected on a single grave, to complex settings in an enormous variety of shapes (Bennett 1987). The latter include kerb rings, circles, elipses, rectan-
33 – Death Ritual and Mortuary Behaviour
877
gles, star patterns, triangles, often with other stones within them forming still further variant patterns. We also find curious three-sided forms with concave sides, known in the absence of an English term as treuddar, ‘three-pointers’. The meaning, if any, of all these stone settings is undetermined, but several explanations have been proposed — by way of example, a recent idea has seen the treuddar as representing the roots of a tree, perhaps yggdrasil, the World Tree (andrén 2004a).
a particularly striking form of stone setting is shaped like a ship, occurring in a range of sizes up to an enormous 356 m long at Jelling in Denmark (Holst and others 2012). The ship settings are sometimes empty but found among graves, often with the remains of fires and meals within — perhaps some form of commemorative place. other ship settings contain one or more cremations spaced around their interior. in general most graves contain single cremations, but multiple burials in the same mound are known and are not uncommon within the larger stone settings. There is also a wider but related issue in the erection of memorials to the dead beyond the burial itself. These will not be treated in detail here but include rune stones, standing stones, bridges, and monumental acts of commemoration such as colossal mounds, fortresses, and churches (see Roesdahl 2005; a.-s. gräslund 2008b).
Uniquely on the island of gotland, memorial stones carved with pictures occur, erected as upright slabs of local limestone up to 4 m high. With a tradition beginning in the Migration Period, the stones evolved through several forms, design schemes, and scales to the end of the Viking age. Their iconog-raphy is too large a subject to tackle here, as the picture stones represent probably the single biggest corpus of late iron age visual thought that has survived, but a complete catalogue was compiled by sune lindqvist (1941–42), updated by Erik Nylén and Jan-Peder lamm (2003).1 outside gotland, the only other such picture stones known are one from Uppland, two from Öland, and one from grobin in latvia, all thought to commemorate gotlanders who died there (Nylén and lamm 2003: 144–47; Petrenko 1991).
Cemeteries
The great variety of funerary groupings reflects the spatial and no doubt social patterning of the communities that they served, from individual farms with
‘family plots’ to larger villages and urban centres such as Birka and Hedeby.
1 subsequent finds of picture stones are reported in the annual volumes of the journal Forn-vännen, and new studies to be found in Herlin Karnell (2012).
878
Neil Price
There also seem to be political factors at work, in that some areas tend to aggregate the dead in clusters whereas others maintain traditions of local burial. in general the dead were not buried far from settlements, but instead their graves can be seen as an extended component of the inhabited areas. as with the revi-sionist attitude to the study of individual graves discussed above, cemeteries themselves should probably be seen less as archaeological site plans and more in terms of landscapes of experience.
The ‘grey literature’ of archived archaeological excavation reports contains evidence from hundreds of Viking age cemeteries all over scandinavia; thousands more have been provisionally identified through topographic or visual survey (e.g., ambrosiani 1964 and Hyenstrand 1974, for central swedish sites).
Until a thorough review of this somewhat neglected resource is complete, we can only skim the surface of the available data. However, a few notable sites can be mentioned.
The best example of a cemetery as it originally appeared is found at lindholm Høje in northern Jylland, where a grave-field was buried by wind-blown sand and has survived intact (Ramskou 1976). almost every burial is marked by stones, often without apparent pattern, but clearly comprising an integral part of the funerary ritual. other important and well-published sites include Valsta village in central sweden, with its syncretic funerary rituals blending Christian elements with more traditional rites (andersson 1997, 2005a), specialized communities such as the tenth-century enclosure at fyrkat (Roesdahl 1977), and spectacular individual graves within larger grave-field contexts such as Mammen, also in Denmark (iversen and others 1991). Regional surveys have been produced for the area around göteborg (B. Hall 2007), and langeland (grøn and others 1994) and Vestjylland in Denmark (Eriksen and others 2011: 149–76).
Urban cemeteries form a discrete category of their own, notwithstanding the endless debate on the definition and character of such settlements.
Providing merely a representative selection, at the larger end of the scale a number of excellent reports have appeared on three of the most important sites: namely Birka, Hedeby, and Kaupang.
The late nineteenth-century excavations in Birka’s cemeteries (published as arbman 1940–43) have spawned several major monographs and countless smaller works using the objects from the graves as a foundation for the study of Viking age material culture in general, at least in sweden. However, apart from anne-sofie gräslund’s monumental work (1980), less subsequent attention has been paid to the rituals in evidence there. This is now starting to change with major projects from both stockholm University and the National
33 – Death Ritual and Mortuary Behaviour
879
figure 33.8. aerial view of the burial ground at lindholm Høje in Vendsyssel in northern Jylland, dated to the late iron age. The cemetery, which is one of the best preserved from the iron age, consists of standing stones, many placed in different geometrical forms, such as squares, circles, triangles, and ovals. Photo: Nordjyllands Historiske Museum, aalborg.
Heritage Board, focussing on the so-called Black Earth and the burials in both the hillfort area and the sectors of the island outside Birka’s wall, in addition of course to the great grave-fields of Hemlanden (summarized in several papers in Hedenstierna-Jonson 2012b, with reference to the bulk of the theses and dis-sertations dealing with the island); several aspects of this work are discussed in more detail below.
like those of Birka, the Hedeby cemeteries also contain mixtures of cremations and inhumations, including some spectacular chamber graves. as geora-dar survey adds to our knowledge of their extent, it is more clear than ever that
— again like Birka — the burial fields are divided into specific zones and that there is a clear funerary topography for the different cemeteries. after many years of disparate publication in papers and report fascicules, discussion of the Hedeby graves has now been gathered for the first time in two volumes (arents and Eisenschmidt 2010). in Norway, the cemeteries at Kaupang are different again, lacking the chamber graves of the other major towns but instead having a greater number of boat graves and similarly spectacular interments. The burials are distributed along the higher ground behind the settlement at the water’s edge, with more graves on adjacent islets and promontories. The publications
880
Neil Price
of the 1950s excavations (Blindheim and Heyerdahl-larsen 1995; Blindheim and others 1999) have now been supplemented by new research from more recent fieldwork (stylegar 2007).
The grave-fields of several Baltic islands have also been published in detail, with especially good coverage of gotland (Thunmark-Nylén 1995–2006)
— including the cemeteries serving proto-urban trading sites (e.g., Carlsson 1999) — and Öland (Beskow sjöberg and others 1987–2001). as we have seen above, similarly sized cemeteries in the overseas colonies are found especially in the east at sites such as staraja ladoga on the Volkhov, and gnezdovo near smolensk (androshchuk 2008, 2013), but here scandinavian burials are inter-mingled with those of other groups.
Under the Earth
Chamber Graves
Chamber graves are known from the centuries before the Viking age, especially in the Roman iron age and Migration Period, but it was in the ninth and especially tenth centuries that they reached their zenith. They effectively take the form of underground rooms built in wood, with walls, floors, and raftered roofs, usually covered by a mound (Eisenschmidt 1994; Ringstedt 1997; stylegar 2005, 2014; fischer 2014). There are no standard dimensions, but the graves are always either square or rectangular; the swedish examples average 2 m deep and may be up to 4 m in length. These are substantial structures requiring considerable investment to build, especially in a scandinavian winter.
Chamber graves were apparently reserved for occupants of high status, buried in fully furnished environments packed with objects, animals, food, and drink.
Here too the dead could be laid out on their backs or sides, or even sitting in chairs. The bodies are also occasionally found in coffins on the floor of the chamber, or in beds.
They are most common in sweden, where 111 examples have been found at Birka alone (arbman 1940–43; Price 2013b), while around sixty are known from Denmark and northern germany (Eisenschmidt 1994). The latter examples cluster around Hedeby, and it seems likely that the early towns were epi-centres for the spread of what became an unusual but interregional burial rite (stylegar 2005). in Norway the custom was not as widespread, and no such burials have yet been found at Kaupang (the nearest equivalent to Birka and Hedeby), and on present knowledge chamber graves appear as a primarily eastern and southern phenomenon, as we have seen above. This burial form is also
33 – Death Ritual and Mortuary Behaviour
881
found in areas of scandinavian
settlement or influence abroad,
especially in Russia and Ukraine
where elaborate chamber graves
have been excavated at Cernigov
amongst other sites (Duczko
2004; androshchuk 2008, 2014).
some of the chamber graves
are among the most spectacular
burials known from the Viking
age. Every grave is different, and
many can be reconstructed as
microcosms of local belief and
funerary practice. only isolated
examples of this rich variety can
be given here, but at Hedeby the
burials include a large chamber
with a ship placed on top of it
(Müller-Wille and others 1976),
and the Mammen grave from
Denmark represents what may be
the resting place of a Viking man
of princely rank. Dating to c. 970,
the chamber was built to resem-
ble a hall, with a pitched roof and
sturdy wooden walls, all buried by
a mound. inside was a wooden cof-
figure 33.9. Reconstruction of a chamber
fin-box, on the lid of which stood
grave in Birka (grave 845), with the dead
a candle. The rich textile finds in
woman seated in the grave. illustration:
particular have revolutionized our
Þórhallur Þráinsson in Price 2002: 141.
knowledge of high-status male
dress, and the silver-inlaid axe is among the most famous finds from the whole period, giving its name to the Mammen art-style (iversen and others 1991). The greatest chamber grave of Denmark, probably built by King Haraldr blátǫnn for his father as part of the Jelling monuments, is covered elsewhere (Holst and others 2012).
in some of the chamber graves, especially at Birka, the dead are found to have been buried seated, presumably on chairs or stools, though the latter have decayed. The deceased sometimes have objects placed in the hands or on the
882
Neil Price
lap, with grave goods laid out around and particularly in front of them. in rare examples, as in the tenth-century grave iX at the Vendel cemetery, Uppland, sweden, individuals are found seated in chairs on the decks of ships (stolpe and arne 1912: 37). female seated burials are more common in the chambers, whereas on ships the rite is largely confined to men. Exceptionally in two chamber graves from Birka, men and women have been found buried sitting on top of each other in the same chair, the woman uppermost in both cases (Price 2002: 132–39). Remains of slim iron chains around the bodies suggest that the corpses were tied to the back of a chair to hold them in place.
The meaning of seated burial is not known, though it is clear that in at least some instances the graves have been deliberately oriented so that the dead seem to ‘look out’ over a specific vista. at Birka, for example, the chamber graves with seated women are all positioned so that their occupants’ faces would be turned inwards to the town, perhaps watching over it (Robbins 2004). We should also note that seated burial is mentioned in a different kind of source, ibn fadlan’s eyewitness account of the Volga ship cremation. Here, cushions are used to prop up the dead chieftain’s body in a sitting position on top of a bench that has been made up as a bed.
Boat and Ship Burials
stone settings in the shape of ships have been mentioned above, but the most spectacular burial rite of the Viking age involved the deposition of actual ships in the graves (Müller-Wille 1970). in burials like this, vessels ranging from dug-out canoes and small fishing skiffs to thirty-metre longships were dragged up from the water and laid to rest in a trench specially cut for the purpose. The bodies of one or more men and women were placed on board, laid out in various ways: lying amidships or resting in bed, sitting in chairs, or propped up on cushions, sometimes covered by shaggy bearskins. The dead were often deposited in a chamber, usually built amidships but occasionally underground with the ship balanced on top (sometimes inverted). as in other chamber graves, they are found surrounded by weapons and personal belongings, jewellery, tools, household equipment, and fine textiles. The makings of a feast were often laid out — decorated drinking horns and expensive imported glasses, and the mead and wine to fill them; alongside them are offerings of food, herbs, and spices. sometimes the full range of home furnishings was carried on board: beds, looms for weaving, smiths’ tools and agricultural equipment, sledges, ice skates, even entire wagons. Tents and other gear for the outdoor life have also been found.
33 – Death Ritual and Mortuary Behaviour
883
figure 33.10. Reconstruction of the spectacular final phase of the burial at oseberg in Vestfold in 834, envisaged to have taken place at night. illustration: anders Kvåle Rue.
a second category of ship graves involves the burning of the vessel, as in the account of ibn fadlan. in such cases the ship was lifted onto a great pyre of wood and set on fire. We find it as a heap of ash with the outline of the vessel preserved as rows of iron rivets, filled with burnt and twisted traces of what had lain inside.
in sweden, ship burials cluster in the Mälar Valley, especially at the site of Valsgärde, which has a continuity of boat graves at a rate of one per generation since several centuries prior to the Viking age (arwidsson 1942, 1954, 1977; lamm and Nordström 1983; Munktell 2013). other major sites include Tuna in alsike (arne 1934) and Vendel (stolpe and arne 1912), while significant boat burial excavations outside the Mälar region include Klinta on Öland (Petersson 1958, 1964) and skamby in Östergötland (Rundkvist and Williams 2008).
Danish ship burials are fewer in number but no less dramatic, including the remarkable grave from ladby (sørensen 2001) and the example from Hedeby (Müller-Wille and others 1976). Here the tradition of boat burial has its origins earlier in the iron age and may offer clues as to the significance of the vessels in that parts of ships were buried with the dead in the absence of the complete craft (as at slusegård: andersen and others 1991).
884
Neil Price
The most dramatic examples of the ship burial rite have been found in Norwegian Vestfold, with the famous burials at oseberg, gokstad, and Tune (Nicolaysen 1882; Brøgger and others 1917–28; Christensen and others 1992). Due to their unusual degree of preservation — protected by the anaero-bic clay of their mounds, which has left not just the vessels themselves but also organic grave goods intact — these burials are among our richest sources for the detailed inventories of high-status graves anywhere in the Viking world.
other major boat burials include several from gausel (Børsheim and soltvedt 2002), avaldsnes (opedal 1998), and Kaupang (Blindheim and Heyerdahl-larsen 1995; Blindheim and others 1999).
Beyond scandinavia, boat burials are found in the British isles, especially in island communities on the orkneys (owen and Dalland 1999) and Man (Bersu and Wilson 1966). These burials are sometimes lined with stones in the prow and stern (graham-Campbell and Batey 1998: 135–40), as at ardnamurchan (Harris and others 2017). Beyond Denmark, until recently there was only one scandinavian ship burial known from Continental Europe, located on the Île de groix off the south coast of Brittany (Müller-Wille and others 1976; Price 1989), but this has now been joined by the extraordinary twin ship burials found at salme on the Estonian island of saaremaa, discussed below in the section on war rituals.
The dead were often accompanied by very high numbers of animal sacrifices
— up to twenty decapitated horses, for example, accompanied the oseberg grave. Whole or partial bodies have been found of creatures ranging from domesticated livestock such as cattle, sheep, pigs, and goats, to riding horses and high-status beasts such as hunting dogs, falcons, and several species of hawk. Even more exotic creatures are sometimes found in the ships, including owls, peacocks, eagles, and cranes, to name but a few. in rare instances, fish are also found in the boat graves.
Regardless of the nature of the rituals involved, almost all the graves are covered with a mound of earth, a barrow erected over the buried vessel or the remains of its cremation (the Uppland boats are a notable exception, though they too may have various forms of aboveground marker). sometimes a post might be set up on top, the ship’s mast left sticking vertically out of the mound, or the beaks of the prow and stern left exposed out of the barrow’s sloping sides.
The burial monuments themselves can be augmented with other features, such as the circle of standing stones surrounding the groix ship grave, and the line of stone uprights that appear to form a processional way leading up to it (Müller-Wille and others 1976).
in boat burials too we find regional variation, sometimes startlingly so as in the case of the island of gotland. No ship graves have been found on the
33 – Death Ritual and Mortuary Behaviour
885
island, but instead Viking age (and earlier) burials are sometimes marked by picture stones (see above), that in their lower sections usually bear a depiction of a ship under sail. it has been suggested that these picture stones are in effect the gotlandic equivalent of ship burials, but with their message content expressed through images rather than the physical objects that are customary on the mainland (andrén 1993).
The content of this message and meaning has been subject to long debate, focusing principally on the ship as means of transport for a symbolic journey or as a high-status possession either of the dead or their wealthy relatives (Crumlin-Pedersen and Munch Tyhe 1995). others have seen the burial ships as metaphors, representing the hall buildings characteristic of high-status chieftains, and thus forming a kind of residence for the dead (Herschend 1997). in this interpretation it is important that the dead stay in the mound, protecting or serving their community with spiritual power — the very opposite to the idea of death as a journey.
We certainly know that some of the ship burials were left partly open for a short time, with portions of the vessel visible. The most striking evidence comes from the oseberg ship burial, which has been shown to have been covered only partway by the original mound, leaving the entire prow and fore-part of the ship exposed, including the entrance to the burial chamber (gansum 2004; Nordeide 2011c critiques this interpretation, but shows only that the grave must have been ‘completed’ in the course of a season — the sequence and staggered nature of the rituals is confirmed). although the mound was later extended to cover the whole vessel, we do not know what kinds of activities took place around and even inside the burial in the intervening period. similarly, the line of boat burials at Valsgärde may have been housed under open-ended sheds, their occupants and contents accessible to visitors (Herschend 1997).
While these questions are not easy to resolve, it is clear that the ships often contain deliberate markers of ethnicity, religion, and power, and may also hold the clue to remarkable cultural interchange. one example comes from the Uppland graves as a whole, in which the presence of sámi objects has been found in some profusion, including entire sheets of decorated birch bark tent covers that seem to have been laid over the ships at both Vendel and Valsgärde (Price 2002: 237). DNa and dietary work at the Tuna in alsike grave field has also suggested that the dead interred there may have had sámi ancestry (Price 2002: 237), raising the question of whether some of the ship burial occupants may actually have actively maintained sámi identities.
similarly startling results were obtained from new work on one of the two women from the oseberg burial. originally thought to have been aged about
886
Neil Price
sixty to seventy and twenty-five to forty years old, respectively, when they died, analysis of tooth root translucency in the younger woman has shown that she was probably at least fifty and perhaps older still, thus closing the age gap between the two. Most interestingly, successful extraction of DNa from one of her teeth has revealed that she belongs in mitochondrial sub-haplogroup U7, which strongly suggests that she came from the Middle East, particularly the area of modern iran (Holck 2008: 205, 208). Very close matches in radiocarbon dating sequences indicate that the two women most likely died at the same time, while Carbon-13 analysis shows that both women had followed the same diet, perhaps implying that they were of similar status (Holck 2008: 204–05).
The same pathologist’s more recent studies (Holck 2009a, 2009b) suggest varying degrees of genetic abnormality and severe illness in the older oseberg woman and the man from gokstad, giving them dramatically changed appearances and perhaps thereby some kind of marker that made them appropriate for the rite of ship burial.
another striking aspect of the ship burials is their construction for both women and men. it is clear that the latter were in the majority as primary occupants of the graves, almost 90 per cent of those interred being male (Müller-Wille 1970; subsequent work has not significantly altered this ratio). However, the female boat graves are also among the richest known: indeed, the two women of oseberg occupied the most spectacular Viking age grave ever found (though on the basis of the artefactual assemblage, one scholar has argued that the primary burial at oseberg was actually that of a man, whose body was completely removed when the chamber was disturbed; see androshchuk 2005). Even in a minority, these rituals have considerable implications for the status of women in Viking society and accords well with other female-sponsored memorials such as the runic inscriptions mentioning bridge-building and similar activities. it is also worth noting that many of the boat graves are situated in cemeteries with other types of burials, as at Valsgärde where cremations are interspersed among the boats and include some female graves of very high status.
Old Graves, New Dead
another phenomenon that emerges clearly in Norse mortuary behaviour in several areas of the Viking world is a deliberate concern for the relative location of the dead, not only regarding proximity to the farms of the living (cf. adolf friðriksson 2013) but also to the burials of much earlier times. in our reconstructions of ancient world-views it is easy to forget that the people of the past themselves had a prehistory (Bradley 2002; andrén 2013a). The perceptible
33 – Death Ritual and Mortuary Behaviour
887
figure 33.11. Raknehaugen in Ullensaker in akershus. The grave mound is one of the largest in scandinavia and has been dated by dendrochronology to 552. it covered a cremation without any objects deposited in the grave.
Photo: Bjørn Haugen, Kulturhistorisk museum, Universitetet i oslo, oslo.
cognitive distance to, for example, the Bronze age was scarcely less for the Vikings than it is for us, and while their understanding of their own past may not have been as ‘accurate’ as ours, there is no reason why their explanations of it were any less sophisticated. in particular, the siting of Viking age burials close to prehistoric barrows and mounds may relate to ideas about the ancestors (laidoner 2015), and the social situation of contemporary tenants in relation to the age-old inhabitants of the land.
Thus in Denmark and southern sweden, there are numerous instances of Viking age graves being inserted as secondary burials into spectacular mounds from the Bronze age, or alternatively created around them in the landscape (Pedersen 2006; see also Thäte 2007 for a broader scandinavian frame of reference). The same phenomenon can be observed on gotland, where Viking age cremations have been carried out in a broad ring around the base of great stone cairns from the Bronze age, as at Uggarde Rojr (Martinsson-Wallin, on-going excavations and personal communication). in similar studies of eastern central sweden, ann-Mari Hållans stenholm (2006, 2012) has shown how intricate
888
Neil Price
figure 33.12. Double inhumation grave
from the eighth century in Birka. The grave
contained two men: at the bottom an
older man with weapons (B) and on top of
him a beheaded younger man (a) without
any objects except for an antler of elk. The
young man with the elk antler has been
interpreted as a human burial sacrifice.
after Holmquist olausson 1990: 176.
sequences of superimposed graves have been used to shape formalized spatial relationships to the past. often there is an immense gap in time, some thousand years or so, between the earliest graves and the ones overlying them: from the Roman iron age to the Viking period. in a variant of this practice, we also find some of the so-called storhögar (‘great mounds’, a term used to denote barrows of high social status and monumentality) constructed as successive layers of burials, built up over time throughout the later iron age as they acquire new levels of funerary meaning within the community (Bratt 2008: 62–97; see also gansum 2004). similar practices can also be found within the Viking age itself, as in the secondary deposition of cremations into existing inhumation graves (especially in the swedish province of skåne; see svanberg 2000). This presumably functioned as part of similar processes of commemoration as we have seen above, binding people with the land and turning the varied monuments to the dead into what have been called ‘memory machines’ (Jones 2007;
33 – Death Ritual and Mortuary Behaviour
889
see also Williams 2006). The location of burial mounds, especially on boundaries, is used as a signal of tenure certainly within anglo-saxon juridical practice (semple 2013), and in scandinavia the siting of barrows is also found in the legal documents of much later periods as proof of long-term rights to the properties in question (Zachrisson 1994).
Death at the Funeral
Human sacrifice in association with burial can be hard to identify with certainty, as graves with more than one occupant may represent family groupings or multiple burials due to disease, amongst other possibilities. However, a significant number of Viking age graves contain individuals who were clearly killed to accompany the primary occupant of the burial in death — diagnostic injuries in these cases include decapitation, stabbing, broken necks, and hanging, with the hands and/or feet sometimes being bound.
famous examples include a man buried at stengade with a decapitated, bound man placed beside him, both bodies covered by a heavy spear (skaarup 1972), and a similar burial from the hillfort wall at Birka, in which the decapitated body of a young male was laid partly over that of an older man furnished with weapons and with elk antlers placed behind his head (Holmquist-olausson 1990). another tied, decapitated man accompanied the male buried at lejre (andersen 1960), while a woman’s grave from gerdrup near Roskilde contained the body of a man with a broken neck (Christensen 1981). at Ballateare on the isle of Man, an armed male youth had been buried with grave goods and covered by a mound, on top of which a young woman was killed with a sword blow from behind, apparently while kneeling. a thick deposit of cremated animal bone was then strewn over her body, covered in turn by a second layer of earth to complete the mound (Bersu and Wilson 1966).
The human accompaniment of the dead seems to have been particularly common in connection with ship burials. The most dramatic case comes from the account of ibn fadlan mentioned several times previously (Montgomery 2000). The ship cremation ceremony (è32) includes the murder of a young slave girl — the arabic implies that she was about fourteen or fifteen years old
— stabbed and strangled after at least six acts of rape and many more of arguably consensual sex. During the course of the rites she is seemingly drugged with some sort of beverage and has (or says she has) a series of visions. ibn fadlan states specifically that the girl volunteers to accompany her owner in death, though how much coercion was involved is another matter. He mentions that slaves of both sexes might do this, and also dead men’s wives and concubines.
890
Neil Price
The latter are also mentioned by other arab writers such as ibn Rustah and ibn Miskaweih, who describes how women might be buried alive in the chamber graves of their male partners, something perhaps confirmed by the Russian burials at Cernigov (Price 2002: 46). it is clear that more than one person might be sacrificed at Viking funerals, and we have seen above the Byzantine account of a Rus army burning its war dead by moonlight, accompanied by the mass killing of prisoners of both sexes (Price 2002: 369).
studies of Norwegian burials containing more than one person, in circumstances that might imply sacrifice, have found that many of the presumed victims had markedly different diets from the ‘primary’ occupants of the graves, though similar to that of the majority population (Naumann and others 2013).
This does not confirm (or disprove) that the sacrifices were slaves, but does suggest a difference in status between them and the other persons in their graves.
War Rituals
some late Vendel and Viking age burials relate to circumstances far removed from ‘conventional’ funerary settings, and instead represent the treatment of casualties sustained in battle. Buried far from home and in situations not necessarily of their comrades’ choosing (for example, under pressure in hostile territory), these give us brief glimpses of what might be termed war rituals, a focused concern for the martial dead, the manner of their passing, and the form of their memory.
Until recently, the most dramatic of these was the large complex of burials and fortifications around Repton church in Derbyshire, associated with a winter camp of the Viking great army and closely dated to 873–74 (Biddle and Kjølbye-Biddle 1992, 2001). The church had been used as a gateway strong-point into a D-shaped enclosure on the river bank, with several burials in pre-Christian scandinavian style around it. These included a clear warrior grave of a man felled by several massive injuries and buried with weapons and pagan charms (a Þórr’s hammer, jackdaw’s leg, and a boar’s tusk between his legs), with a second grave adjacent that contained the unfurnished body of a youth; both graves had been covered by a stone cairn through which a standing post had been erected. outside the fortification, a former mausoleum of the Mercian royal line had been adapted for use by the Vikings, with a huge charnel deposit of disarticulated bones — representing more than 250 people, mostly men —
arranged around what is thought to have been a central burial. after years of debate, adjusted radiocarbon dates and isotope analyses have confirmed that the bodies are indeed related to the Viking great army.
33 – Death Ritual and Mortuary Behaviour
891
Two mass graves of what appear to be execution vic-
tims have been excavated at Ridgeway Hill, Weymouth
(loe and others 2014), and st John’s College, oxford
(Wallis 2014). all the bodies are male adults, and iso-
tope data indicate a scandinavian origin for many of
them, but there is no question that they were killed
and buried by their enemies, and thus do not repre-
sent properly scandinavian rituals. There is, however,
a degree of ceremony involved, especially at Ridgeway
Hill, where the decapitated bodies were thrown into
a pile, with a stack of heads adjacent, all in a disused
Roman quarry on prominent high ground.
interestingly, perhaps the most significant Viking
funerary discovery of the last century also belongs to
this category of war rituals, in the find of two boat bur-
ials dating to c. 750 at salme on the Estonian island of saaremaa. Excavated between 2008–12, these remarkable graves appear to represent the resting places of
scandinavians who perished on a raiding expedition.
The find is unique and gives us an unprecedented
insight into what the ships, crews, and material cul-
ture of such forces might have looked like at the very
start of the Viking age. as this essay goes to press,
the salme finds are still under investigation, and this
account is based on interim publications (Konsa and
others 2009; Peets and others 2011, 2013; Peets 2013;
Curry 2013), but they undoubtedly represent one of
the most exciting early Viking discoveries for many
decades.
originally located on a spit of sandy terrain project-
ing out into a heavily trafficked sound between islands,
two clinker-built boats had been partly buried so as
to leave most of their profile above ground. aligned
parallel to the shore and only a few metres from the
water, they would have been visible from a distance
figure 33.13. Reconstruction of the burial of seven men in the small boat grave at salme on saaremaa (Ösel) in Estonia. illustration: Þórhallur Þráinsson.
892
Neil Price
and must have remained a landmark for sailors over many years. The sea would have washed their hulls, as we know from the lenses of silt that gradually built up against the coastward side of the boats — a dramatic sight.
The two vessels were of different sizes, but from the context they seem to have been deposited at the same time and thus probably resulted from the same event. The salme ships together contained an astonishing forty-one bodies, an enormous number by comparison with any other known grave from the Viking age (even the greatest previously known ship burials had no more than four or five occupants). The dead ranged in age from late adolescence to maturity, with the majority in their thirties, and thus represented men in their prime. Many of the skeletons showed signs of major physical trauma consistent with battle injuries, indicating that these men died in combat; the others presumably perished at the same time, of unknown causes. The fighting had evidently been at least partly maritime in nature, as the sides of both vessels were studded with arrowheads, including a triple-pronged fire arrow used to carry wads of burning material into enemy hulls, sails, and rigging.
The smaller of the two vessels was a rowing boat approximately 11.5 m long and 2 m wide. it contained the bodies of seven men, placed apparently sitting up on the benches: six of them in three pairs at the oars, seemingly with their arms round each other’s shoulders, and the seventh man at one end (we cannot distinguish the prow from the stern). The men were buried with weapons and a range of personal items, and the boat also contained hawks, falcons, and dogs
— all animals associated with high-status activities such as hunting.
The second vessel was much larger, a true ocean-going ship 17 m long and 3 m across the beam. it is hard to be sure from what was preserved, but it seems very likely that the larger of the salme boats was powered by sail. at one end of the boat (again, we cannot tell the prow from the stern), thirty-four men were buried in four layers, laid down side by side in the hull. More swords than men were present in the grave. The quality of the workmanship was very high, with a great deal of gold ornament. The bodies were strewn with a variety of objects, such as gaming pieces; on some of the corpses, fish of several species had been placed. at the centre of the mound was the richest of all the salme individuals
— a man laid down with weapons and jewellery of spectacular quality. in his mouth was a single gaming piece, the king.
The whole pile of bodies was covered by a kind of mound made from fifteen shields, placed with overlapping boards to form a wooden dome over the dead. Each shield boss had been hammered flat, and many of the weapons were deliberately bent. over this lay a large piece of textile, perhaps the sail. in the ship were a number of dogs that had been killed in various elaborate ways, pre-
33 – Death Ritual and Mortuary Behaviour
893
sumably sacrifices to accompany the dead. The fact of the careful and elaborate burial rituals surely implies that the dead were interred by their countrymen, and the undisturbed nature of the find suggests that the scandinavians retained sufficient influence over the region to prevent the desecration of the graves.
Concluding Remarks
after decades of study with an ever-increasing data-set, it has become clear that the mortuary behaviour of the later scandinavian iron age saw a gradual development of two trends by comparison with preceding centuries: the broad standardization of outer burial form, and an almost infinite variation of the individual rites practised within that repertoire. The key questions in the study of deathways at this time focus on the possible explanations for this transformation. Beyond their intrinsic study, these problems can also be illuminated by comparative reference to the northern germanic societies of the Continent, the burial rites of the anglo-saxons, and the rituals of the sámi.
a necessary first step here is the deconstruction — at times a fantastically difficult task — of the most apparently fundamental terminology: put simply, in the later iron age, what was a grave, and what was a funeral?
it is clear that the rituals involved were complex affairs and took a considerable period of time, sometimes so long that the body was placed in a temporary grave while the primary site was being prepared (an example of this has been excavated in northern iceland; Roberts 2009). ibn fadlan notes that the funeral rites took a full ten days of continuous activity, which may not be untypical.
good evidence for these drawn-out rituals of death comes in particular from ship burials, such as a boat grave from Kaupang that exhibits a particularly prolonged sequence of activity (Price 2010, 2012).
The chamber graves in particular also exhibit a complexity that must reflect an intricate series of actions during their construction, such as the burials of possible sorceresses on Birka (Price 2002: 128–41). one of these, Bj.834, contains a double chair burial as described above, and a lance has been thrown across the seated figures in order to strike deep into the wood of the platform upon which rests a pair of draught-harnessed horses. other burials also exhibit weapons being either stuck into chamber walls or else plunged vertically into cremation deposits (Nordberg 2002).
one further element of this extended funerary behaviour may be the practice of so-called grave robbing. While clearly some burials were merely plundered for their valuables, many of the break-ins to mounds and other graves are so extensive that they simply cannot have been done in secret or without
894
Neil Price
wider social sanction — the disturbance of the oseberg and gokstad burials is a case in point (Bill and Daly 2012). often burials were opened (perhaps a better term than ‘broken into’) soon after the original interment, as seen in the still partial articulation of the corpses when they were disturbed. While some of these removals have a relatively clear motive, such as the translation of gorm the old’s bones to the new church at Jelling, others are more obscure. often the bodies are moved around or taken out altogether, some objects are taken while others are left alone, and sometimes it is possible to see how piles of items were shifted en masse and left where they were placed, presumably in order to access something else. Excellent work has been done on this phenomenon in an anglo-saxon context (Klevnäs 2007, 2014), and it is clear that further research will shed much-needed light on an integral part of the mortuary behaviour that has hitherto been erroneously considered only in relation to the actual burial itself.
There is a strong sense of the active presence of the dead in the grave. in some of the Valsgärde ship burials, for example, hnefetafl sets are laid out with the pieces positioned as if in the middle of a game, with the next, winning move due to the player on the side of the board nearest the body of the deceased: the dead man is about to beat his invisible opponent (Herschend 1997). in the princely chamber grave from Mammen, Denmark, a candle had been lit on top of the coffin, and would have burned down, unseen in the dark, as the oxygen ran out in the sealed grave (iversen and others 1991). Then there are all the
‘missing’ graves. We still know very little of the criteria that decided whether any given individual received a burial at all, but equally we should not assume that an archaeologically invisible funeral was necessarily one without dignity or meaning. Perhaps they too were ‘present’, but somewhere else.
another important dimension of the grave was the role played by funerary ritual, and its interplay with burial form, in constructing a posthumous identity and memory for the dead. interesting studies of runic epitaphs for men has shed light on the desirable components of a masculine memory, coupled with the qualities conferred by specific kinds of deaths (Thedéen 2009). This has considerable implications for what might be meant by the selection of particular kinds of objects for burial with the dead — for men, an obvious example being the clichés of ‘warrior graves’ containing weapons. The choice of what has been put in, or left out, may contain subtle messages about the reputation and character of each individual, deliberately constructing or obscuring aspects of their status after death. individual objects, such as box brooches, may be employed as signals of ethnic allegiance — they are conventionally associated with female dress of gotlandic type — but may equally form part of intricately cosmopolitan identities (Thedéen 2012).
33 – Death Ritual and Mortuary Behaviour
895
We must also consider whether graves were a success, in contemporary terms.
The concept of ritual failure has recently begun to be explored in archaeology (Hüsken 2007; Koutrafouri and sanders 2013), and this is a perspective that can also be profitably applied to burials. in assessing the social impact of late iron age funerals, we often routinely speak of costly and conspicuous consumption in the choice of objects and animals disposed of in the grave: for example, horses were expensive creatures, and the slaughter or more than a dozen in some of the larger ship burials must have represented a considerable sacrifice in economic terms. This has been interpreted as signalling the high status of the dead, or their relatives, or the community within which the rituals are taking place — but even assuming that we are right in sensing the presence of such markers, what is to say that any of these social strategies worked? among the onlookers at a spectacular funeral, might there not have been some who snig-gered at the social pretensions on display, and might the deposition of so much obvious wealth been perceived as vulgarity rather than grandeur? We will never know, but we should be open to the possibility, and to the unbidden elements of subversion that can be present at any public occasion.
it is also possible to pursue mortuary behaviour beyond the dead themselves. one example of this is the phenomenon of hoard deposition. it has long been clear that buried hoards of silver and other metals are too numerous for them to represent nothing more than primitive banking, the Viking age equivalent of hiding one’s money under the mattress. given the very large numbers of hoard finds within relatively small areas, especially gotland, it is similarly evident that those doing the burying cannot all have died without telling anyone else where their wealth was concealed. There were probably many concurrent explanations for hoarding behaviour, but it is possible that in part it could relate to mortuary ritual either in the absence of a corpse or in addition to one disposed of elsewhere. There is also an alternative, relating to the actions of a person in advance of their own death. We know that some ambitious individuals were capable of erecting runic memorials to themselves in their own lifetimes, and should therefore reconsider snorri’s suggestion — mentioned above
— that hoarded wealth could be buried by the person who had accumulated it in order to enjoy it in the afterlife. scholars have often been too ready to dismiss details of the Ynglinga saga account, and yet this is the kind of telling observation that is at least as likely to reflect Viking age reality as it is snorri’s imagination (gruszczynski 2019). it is also entirely possible that something similar lies behind the numerous depositions of precious metalwork in watery, liminal locations such as bogs, tidal zones, and rivers (see lund 2009 for a comprehensive discussion of this phenomenon).
896
Neil Price
one obvious factor in understanding Viking burials is the question of where the individual dead were thought to be destined. We know relatively little of specific afterlife beliefs, and what we do know contains many contradictions.
The graves give us small clues, though hard to interpret. The buried dead, and even the accompanying horses, sometimes wear crampons on their feet — does this imply that the funeral took place in winter, or are the dead travelling to somewhere cold? The written sources mention special ‘Hel-shoes’ to speed the dead on their way (strömbäck 1961); is this something similar? one of the bodies in the boat burial from scar on sanday, orkney (owen and Dalland 1999), was that of a man whose feet had both been twisted round to face backwards — was this to prevent him from following the others on their way, or to stop him coming back to haunt the living?
When vehicles are involved, especially ships, it is often assumed that death was therefore a journey, and that the deceased would travel by boat, wagon, or sled into the next world. This may be true, but there is no reason why these might not just represent exceptionally expensive possessions of the dead (or their living relatives) alongside all the other artefacts. in the greatest ship burial of all, oseberg, the vessel was actually moored in the grave by a hawser tied to a massive boulder; apparently the intention was that it should not ‘travel’ anywhere at all. There are also indications that the larger of the two salme vessels was similarly anchored to the ground (Jüri Peets, personal communication).
The funerary symbolism of doors and openings has also been considered by Marianne Hem Eriksen (2013; cf. andrén 1993), with something of a hallmark of current studies in its emphasis on plurality, nuance, the importance of context and also the empowering aspects of gender across a broad spectrum. The future for Viking funerary studies has great potential, but also equal risk. We must learn to embrace diversity, to resist easy categorization, and to appreci-ate each burial for what it was — something as enormous as the ending of a life, unique for each person, and perhaps the beginning of a different one. in the image of the threshold it may be that here we have the ultimate metaphor for the period: the passage of its inhabitants from one place — one life — to another, through a portal that was once wide open but is now closed.