Archeology has long been the Holy Grail of research into the conversion of Scandinavia to Christianity about a thousand years ago. The written sources are laconic and biased, so they cannot be used to form a reliable overall image of the conversion. Scholars therefore have looked to archeology to provide more and better information, especially as the archeological material (unlike the written sources) constantly expands when new discoveries are made. The source basis for the book under the review is impressively large: records of almost 500 excavated burials, in addition to urban locations and remnants of pagan and Christian cult sites.
An additional advantage with archeological sources is that, as the old adage goes, the spade does not lie (unlike the tendentious authors of written sources). But the famous retort still applies: the spade does not lie, because it is mute. Archeological sources do not yield any useful information until they are put into an interpretative framework that extracts meaning from them. In this respect, Nordeide's book often falters.
The stated objectives of the work are two. In the first place, Nordeide wishes to "improve the chronology of the Christianization process...in Norway based on new analyses of archeological sources." The second objective is "to improve our knowledge of the local population, and at what point they left their former religion behind" (18-19). Both objectives concern dating, and in this respect they are poorly chosen research questions, since the source material used is too difficult to date to allow Nordeide to contribute a more exact chronology of the Christianization of Norway.
Nordeide is an archeologist, and she has herself participated in some of the digs that form the basis for her research, although the book is mostly based not directly on the excavated material but on reports from earlier finds, many dating back a century or more. To make the material more manageable, she selected 21 among Norway's 430 municipalities in such a way that all the historical regions of southern and central Norway are represented. This is a reasonable way to deal with the enormous material available.
The first three chapters lay out the background: previous scholarship on the subject, some methodological considerations, and a potted history of Christianity since the time of Jesus. Chapters 4 and 5 contain the analytical part of the book, while Chapters 6 and 7 interpret and conclude.
The long (124 pages!) Chapter 4 treats the relevant archeological finds in the 21 municipalities in turn and in great detail. The chapter illustrates some of the pitfalls in interpreting archeological material. The first region to be treated is Trøndelag, which may serve as an example of the author's approach. The archeological findings that Nordeide summarizes only provide the vaguest suggestions about the progress and chronology of conversion. Thus, she puts them into an interpretative context derived from the Icelandic sagas (the earliest of which are from the twelfth century and thus not good sources for what happened at least a century and a half earlier). The archeological material, Nordeide concludes, "support[s] the information provided by the saga narratives" (126), but this is true only in the most general sense. The archeological site at Mære, for example, contains remnants of an early wooden church under the stone church that still stands. Before the first church, another structure was there, as evinced by postholes. That structure appears to have burnt down, since the postholes are sealed by a layer of charcoal on top of which the wooden church was built, apparently shortly after the earlier structure burnt. That earlier structure is associated with a large number of "gullgubber," thin golden foils with stamped human figures, which are usually (and surely correctly) interpreted as testimony to pagan cult. In Mære we thus have, in Nordeide's reasonable interpretation, an example of a pagan cult site being Christianized by the construction of a church. The question is now when pagan cult ceased and when Christian cult began. This is where the argument becomes muddled. The tower of the stone church fell down in 1270, according to written evidence, so the stone church must have been built before that year. Postholes of the wooden church contains sixteen bracteates struck during the reign of King Sverrir Sigurdsson (1177-1202). The excavator of the site, H. E. Lidén, proposed that the coins were deposited there when the wooden church was pulled down after the stone church had been constructed, which thus could not have been finished much earlier than around 1200. Why the coins must have been deposited in the postholes when the wooden church was pulled down and not when it was constructed remains unstated in Nordeide's text (valuables in postholes are typically interpreted as being deposited there at construction). If the wooden church was thus pulled down at some point around 1200, when was it constructed? Since its construction rested directly on the ground, it would have decayed rapidly. Nordeide concludes that "it is reasonable to date the construction of the church to the end of the eleventh century" (111), a date that a dozen pages later has become "1050-1075" (123). Exactly why we should assume a lifespan of somewhat more than a century for the first, wooden church is never stated. The chronology stated with little reservation for the three consecutive buildings at Mære is highly speculative.
Here, the strong gravitational pull of the sagas is evident. The sagas about St. Olav claim that he was killed in 1030 by Trøndelag pagans, and Nordeide finds those pagans in Mære, where they may be assumed to have continued their pagan practices for a few decades afterwards. The archeological evidence on its own provides only the vaguest of dates for when pagan cult was replaced by Christian cult at Mære, but Nordeide makes the chronology of the site consistent with the saga narratives by arbitrarily assigning a date to the construction of the first church and then, also arbitrarily, equalling that date with the end of pagan cult.
Nordeide uses similar sleight of hand to make the chronology of the site at Hove in Åsen (still in Trøndelag) consistent with the sagas. Most scholars agree that this was a pagan cult site. The only more exact date given in the book is a radiocarbon date of CE 160 ±90 (obviously of no relevance to the end of pagan cult) and most of the datable samples also come from the Roman period, "but they showed continuous activity at the location until c. 1000" (105). This fits very well with the sagas' claims that Olav Tryggvason worked on converting Trøndelag before his death in 1000, so the pagan cult site in Hove "should" have been closed down just before 1000. In reality, the archeological date for the cessation of "activities" at Hove is very vague and approximate; of some twenty-five calibrated carbondates given in the article by Oddmund Farbregd that Nordeide bases her text on, only one clearly postdates the year 800 (CE 945±85). This is not quite showing "continuous activity...until c. 1000." Again, the strong impression conveyed is that the master narrative of the sagas has been allowed to define the interpretation of the archeological material, which is made to appear more exactly datable than it is.
Most of the book focuses, not on pagan cult sites, which after all are rare, but on excavated burials, and whether they are Christian or pagan. How does Nordeide recognize a Christian grave as distinct from a pagan one? "The Christian grave," she writes, "was oriented east-west and inserted into the ground under a flat surface. The body was not cremated, and grave goods were not normally included. The body could be wrapped, put in a coffin, or supported by headstones, and sometimes charcoal was used in the grave" (44). On the basis of this definition, Nordeide identifies each grave as either Christian or pagan, and she then includes these observations in her discussion of "the latest reliably dated non-Christian graves/Norse cult" in each of the 21 municipalities that she has studied. For almost all of them, it turned out to be impossible to come up with any more exact date. The municipality of Gulen may serve as an example: Nordeide examined reports of fifteen graves, fourteen of which could not be more exactly dated than to "the Viking Age." The fifteenth contained an axe, which was typologically dated to 850-950 (162). It is on the basis of such material that Nordeide concludes that "continuous non-Christian cult existed until the end of the Viking Age, which was around 1050" (279). That date will surprise no one, but the evidence adduced does not allow such an exact date. The year 1050 is simply the conventional end point of "the Viking Age." More importantly, such argument runs the distinct risk of being circular. Earlier scholars, on whom Nordeide draws, tended to date to "the Viking Age" any grave that they identified as pagan, because they believed that paganism ended in Norway with the end of "the Viking Age." Most of the graves that she uses for her study were discovered, excavated (often unprofessionally by "farmers and construction workers"), and described long ago, so Nordeide is to a great degree left at the mercy of the information compiled by those earlier scholars. When now Nordeide's conclusion essentially is that paganism ended in Norway around 1050, one is left wondering whether this is not simply restating the old scholarly belief about when Christianity was introduced in Norway. Nordeide does, however, believe in her conclusions: "Future excavations with better methods of excavation and recording may confirm my suspicions" (281). Perhaps they will, or perhaps they will not. It is a disappointment that Nordeide's hard work with so much documentation of so many burials have yielded so little in the way of new results.
Another stated objective of the book is to improve our understanding of how the different regions of Norway accepted Christianity in different ways and at different times. The relevant conclusions are summarized in a table on p. 280. Three of Nordeide's 21 municipalities have produced evidence that she interprets to mean that pagan cult there continued for longer than elsewhere. One of them is the cult place in Mære that was discussed above, where the chronology is in fact very uncertain. Another instance is Ullensaker, where a sword datable to the medieval period ("1050-1536") was found in a mound (222, 357), but Nordeide, wisely, does not judge this information reliable evidence for any kind of pagan cult or burial. The third, and final, example concerns "the last pagan," a woman who was buried in Valle (southern Norway) with an axe, five coins, twenty-one weights, forty beads, several fittings (including seven for belts and a horse harness), a comb, and a set of toilet implements. Nordeide judges this woman not to have been Christian, because she was buried in a mound with so many grave goods. Since one of the coins found in the grave was struck at some point between 1065 and 1080, this is the most recent pagan grave she has found in her material. But was she really a pagan? The question goes to the heart of Nordeide's method. We may compare the Valle grave to one excavated in Broby north of Stockholm in Sweden, where a woman was found with a knife, two coins, a silver ring, three weights, a small wooden shrine with a lock, pieces of iron, and a key. One of the coins was struck at some point 1025-1040. [1] The grave goods are not as rich as that of the woman in Valle, but are similarly varied. Using Nordeide's methods, this woman should be a pagan, but runestones at the site have allowed the excavators to identify her persuasively: Her name was Estrid and there can be little doubt that she self-identified as a Christian, for next to her grave stand the cross-marked runestones she and her sons raised in memory of her first husband Östen, who had died on a pilgrimage to Jerusalem but nevertheless was honored with a mound according to the inscription on the runestone. Being buried under a mound is one of Nordeide's criteria for paganism. Estrid raised at least one runestone that contains the Christian formula "God help their souls." If Estrid was Christian, why could not also "the last pagan" in Valle have been Christian? The problems surrounding this burial means that this book has not in any reliable or useful sense been able to show regional differences in the chronology of Christianization.
Graves such as Estrid's make us question Nordeide's methods, which are based on a rather old-fashioned understanding of Christianity and paganism as mutually exclusive and monolithic alternatives in a binary, rather than as points on a continuum. Archeologists and students of religion understand religions as packages of beliefs and practices that are constantly negotiated, meaning among other things that individual beliefs and practices may be accepted separately (as is obvious in the case of Estrid). Nordeide's methods hark back to a more optimistic time that thought we could definitely distinguish Christian from pagan in the archeological material.
Taken in its own terms, this book cannot be considered an unqualified success. It does not succeed in answering its stated research questions of when Norway or its different regions became Christian any more exactly than previous scholarship. The basic problem is that Nordeide chose to study sources that simply are not suitable for answering her questions. The main sources are burials, which in general are impossible to date more exactly, except for the rare cases when they contain securely datable objects (e.g., coins or high-status weapons). For this reason, they simply cannot be used to give greater chronological stability to our knowledge of the Christianization process, especially since Nordeide's understanding of religion and religious change is unsophisticated. As an orientation in some of the most interesting Norwegian archeological evidence for the Christianization process, the book retains its value.
--------
Notes:
1. Lars Andersson and Margareta Boije-Backe, Jarlabankeättens gravplats vid Broby bro (Stockholms läns museum, Rapport 1999), 4.