Pre-Roman Iron Age Scandinavia
PRE-ROMAN IRON AGE SCANDINAVIA
The Iron Age in Scandinavia lasted for about fifteen hundred years and archaeologists have divided it into a number of distinct chronological phases. The Early Iron Age, also called the pre-Roman Iron Age or the Celtic Iron Age, spans the first five hundred years of the period, from 500 b.c. to 1 b.c. It was during this time that a technological revolution took place that brought the Bronze Age to an end. Bronze was replaced by iron in most tools and weapons. Like the use of bronze, the use of iron was introduced from central Europe; but iron, unlike bronze, did not need to be imported. Known as bog-ore or lake-ore, it precipitated in small clumps below the peat in marshy pools and was a readily accessible raw material. Plentiful resources existed in southern Norway, Sweden, and Denmark. The ore contained many impurities and was not of very high quality, but the Scandinavians developed efficient techniques for extracting serviceable iron by smelting it in simple furnaces. As the skill developed over the centuries, so did the complexity of the tools and weapons, until they were comparable to many others made elsewhere in Europe.
NECK RINGS
During the pre-Roman Iron Age, society was organized by rank. Neck rings were a marker of elite status. Large bronze neck rings, especially the so-called crown neck rings and individual rings with transverse molded bands, are a characteristic element of the set of finds dating to the pre-Roman Iron Age. Altogether, there are forty-seven such rings from Denmark with clear provenances. There are also three Celtic rings. The majority of the bronze neck rings are bog finds; a few are dry-land finds, but none is a grave find. All are individual finds, that is, they are found without any associated goods. Some arm rings and simple neck rings have also been found. They are also bog finds and occasionally appear in cremation burials. The looped ring, another traditional object of this period, was made either of bronze or iron. The majority of the recovered objects that have been fashioned in this way are made of iron. Large iron looped rings are known only as grave finds; such rings would presumably have rusted away in bogs. Looped rings, with a few exceptions, are known only from Jutland. Small and large looped rings are contemporary to each other, and can be seen in the large ring hoards.
BOG OFFERINGS
When land was drained for modern farming, a number of bog burial grounds were found. Bog offerings are archaeologically recorded as early as Neolithic times and into the pre-Viking period. These votive offerings or sacrifices included weapons and even warships as well as human bodies, animals, and assorted artifacts. It is postulated, based on the variety and type of offerings, that they were of a ceremonial nature, part of a fertility ritual or a ritual to ensure success in battle. The earliest bog offerings in the Neolithic period consisted primarily of stone and flint weapons. In the Bronze Age, there were more elaborate sacrifices. Collections of personal items and household objects, such as cauldrons, were recovered as well. Weapons—swords in particular—have also been found. Very often the blades of these swords have been bent back or otherwise damaged, and some argue that this was done to represent a ritual "killing" of an enemy. Animals, particularly horses, were also slaughtered as part of the ritual.
HUMAN SACRIFICE
Human sacrifice seems to have become widespread in the first century b.c. Most of the evidence comes from Danish bog areas, where the bodies of the men and women who were killed have been preserved in the acid soils of peat bogs under anaerobic conditions. The skin, hair, and, in some cases, stomach contents of these bodies have been preserved by the tannins in the peat soils. This extraordinary state of preservation has allowed archaeologists to learn details about the clothing, hairstyles, and diet of these people. Tollund man, a body discovered at Tollund, Denmark, in 1950, is one of the best-preserved bog bodies. He was unclothed except for a leather girdle and a leather cap that was laced across his chin. His last meal was gruel. Around his neck was a hide rope with which he had been strangled before being submerged in the bog. Tollund man is now on display in the Silkeborg Museum in Denmark.
Another example is Grauballe man, also from Silkeborg, who was found to have eaten a final meal of porridge containing chiefly barley, oats, and emmer wheat, along with some weed seeds, shortly before he had his throat cut. He was killed sometime in the first century b.c.
CAULDRONS, WAGONS, AND WEAPONS
The Roman Iron Age and the Migration period saw a return to sacrificial offerings consisting predominantly of weapons. From the later part of the pre-Roman Iron Age, between nineteen and twenty-one cauldrons have been recovered from bogs or in graves. Few were located on dry land. The cauldrons found in bogs and those found on dry land are all individual finds. Cauldrons from graves frequently contain a rich set of associated finds with a full set of weapons (sword, shield, and javelin/lance) and gold finger rings.
The find material of this period becomes markedly variegated, and various imported luxurious items enter the archaeological record. Particularly striking are the two large Celtic display wagons from the Dejbjerg bog in Denmark. The remains of comparable wagons are also known from two cremation burials, one from Langå on Fyn, and one from Kraghede in north Jutland. Imported swords are also found in both bogs and graves.
Swords tend to be solitary finds. Two major weapon deposits of this period are located at Hjortspring bog and Krogsbo⁄lle bog. The great majority of the recoverable archaeological wealth was deposited in hoards during two periods: the early pre-Roman Iron Age and the early Germanic Iron Age. The finds of rings from the early pre-Roman Iron Age are usually interpreted as votive deposits. In the Smederup bog in eastern Jutland, a plank-built well was found not far from the place where great quantities of rings were dug up. It is regarded as a votive well and may therefore emphasize the sacred character of the bog. Artifact studies have shown that artifact types deposited in the bogs of one area are not deposited in graves of the same area.
Two artifacts of great importance have Celtic origins. One is the Gundestrup cauldron, a silver bowl with highly realistic embellishments in relief, including a representation of a human sacrifice; it has been suggested that it was used for catching a victim's blood. Another interesting find is the Hjortspring boat, a war canoe that was unearthed on the island of Als off southeastern Jutland. This canoe carried between twenty-two and twenty-four paddlers and is the oldest surviving example of a boat in Scandinavia. It contained deliberately damaged war equipment, including some single-edged iron swords, which were evidently ceremonial offerings. Studies have concluded that this was a religious deposition of the hoards.
A DECLINE IN POPULATION
One surprising aspect of this period is that it has yielded relatively few archaeological remains. Earlier archaeologists, who worked primarily with grave finds, viewed the pre-Roman Iron Age as a regression period and, in some areas, such as Tro⁄ndelag, Norway, it would appear there was virtually no use of iron. This suggests that the population had declined. Although these early centuries remain comparatively obscure, since very few settlements are known from this period, in the 1990s and 2000s, thanks to a change of focus from grave goods to habitation sites, modern archaeological research has been able to contribute tremendously to our understanding of the pre-Roman Iron Age, providing a new picture of society, especially in southern Scandinavia. In fact, settlement development from the Bronze Age to the Early Roman Iron Age now appears to have been continuous. Certainly the climate, which for about two thousand years had been drier than it is now, became both wetter and colder, so that, toward the north, deciduous trees began to disappear and the glaciers began to re-form on the high ground. Investigations of Danish raised bogs have shown that the climate has fluctuated over the past 5,500 years and that these fluctuations lasted for about 260 years. The climatic changes in the final phase of prehistory can be located with great accuracy. A trend toward increased precipitation and lower summer temperatures set in about 600 b.c., just before the transition to the pre-Roman Iron Age. The next fluctuation took place about 300 b.c., and yet another very close to a.d. 0. This climatic deterioration probably affected the efficiency of farming.
LAND-USE PATTERNS
In southern Scandinavia, the late pre-Roman Iron Age was characterized by woodlands that expanded at the expense of open land (pastures, arable land). This may have been caused by a concentration of settlement in permanent farms and villages. This means that the late pre-Roman Iron Age landscape, broadly speaking, was similar to the Late Bronze Age landscape. On a smaller scale, however, it differed in the organization and land-use pattern of its permanent villages: infields with arable fields and meadows around the farms, and outland with pastures and coppiced woods. In general, the transition to the pre-Roman Iron Age in Sweden did not bring about any sudden restructuring of agriculture. The farms were still isolated, with longhouses the same size as they had been during the Late Bronze Age, with room for one extended family.
This was different from Jutland. There, longhouses became much smaller in the pre-Roman Iron Age, with room for only one family household, but with many houses clustered together like villages. In Scania there were no villages prior to about a.d. 500, unlike in other parts of "Denmark." Before that, in the Bronze Age and pre-Roman Iron Age, there were single farms with Celtic fields, probably under shifting cultivation, which slightly later developed into double or triple farms that seem to have belonged to kin-groups rather than constituting true villages. Nucleated villages were first founded between a.d. 500 and 700. Single farms were not established again with any regularity until the Early Middle Ages.
In Sweden, the excavations of the Skrea project in Halland have unearthed a number of large-scale settlements ranging from the Late Bronze Age to the Early Iron Age. While damage from modern agricultural activities has compromised the preservation of some of the sites, there is still a large quantity of information identifying the settlements as large agrarian units. These settlements are located at dry ridges, often composed of glaciofluvial deposits or other self-draining soil types. Those dry areas were used for living and farming. Vast grounds consisting of heavier soil types suitable for grazing and for hay crops surrounded them. In all archaeological work thus far there has been a clear correlation of site type to soil type. These settlements were inhabited for fifteen hundred to two thousand years, some even longer.
A second type of settlement is smaller and more sporadic. It tends to correlate with different landscape zones, however. Some are on small ridges in otherwise wet areas or in areas with relatively small-scale landscapes. The relationship between the two scales of settlements remains unclear.
Another key site for looking at architecture and settlement is located on the tofts of Lilla Köpinge village. It is in southeastern Scania, near the medieval town of Ystad. It was the subject of intensive investigations, along with Stora Köpinge, which is one of the emporium-like market sites founded in the 800s. Each farm appears to have had its own fixed site, on which several layers of longhouses can be found. The longhouses are relatively large: 17 by 26 by 5.6 meters. Their overall area is not much larger than that of longhouses in the Late Bronze Age, but the greater length of the buildings made it possible to house a greater number of livestock. The farms also had some smaller buildings, including sunken-floor huts, which were used primarily for weaving. In Denmark, the first sunken-floor huts do not appear before the late Roman Iron Age. In the Köpinge area, by contrast, there is concrete evidence dating them to the pre-Roman Iron Age.
MOBILE SETTLEMENTS
One of the observations made for this period is that the settlement was mobile and that villages moved from time to time. Over the long term, they may have come to remain in the same place for longer periods. The greater or lesser mobility of the village communities of this period was first revealed with the extensive excavations at Gro⁄ntoft. Gro⁄ntoft is a rural settlement in western Jutland dating from about a.d. 200, and it provides invaluable information on these Early Iron Age farmers. The settlers must have lived in buildings very similar to those of their Bronze Age predecessors, grouped in villages surrounded by fences. The excavations reveal a single "wandering village" in the same resource territory for a period of about three hundred years. Gro⁄ntoft probably housed about fifty people and about sixty cattle, but it is difficult to know how representative this site is of the period. The houses are of three-aisled construction, which is found at all the Danish Iron Age settlements. This construction dates as far back as the middle of the second millennium b.c. In about 500 b.c., it evolved into a rectangular house shape unvaryingly oriented east-west, with a roof supported by two parallel rows of interior posts. Entrances were found in both long sides of the house. The walls of the houses were sometimes made of massive or light timber and with wattle and daub. There were sometimes also massive earthen and turf walls. The houses were often divided into two sections: the east end sheltered cattle while the west end with the hearth was for human dwelling. The dwelling section often had a clay floor while the barn may have had a stone-paved gutter and stall partitions. At all stages, the village economy strongly emphasized animal husbandry. Houses without stalls did exist, however. When an individual house went out of use, it was torn down and moved to another site within the village territory. The old site was plowed over and the soil was again tilled. The constant moving shifted the original field boundaries marked by balks. The balks (forming the so-called Celtic fields), which were visible at the excavations and thus stem from many phases of cultivation, may have been separated by land left fallow for a period of time. There is evidence of fences dating to roughly 300 b.c.; these were probably used to protect the village and the houses from the cattle.
There are other signs that rural settlements were increasing in number and size toward the end of the Early Iron Age. Many of the Danish settlement sites were excavated in the early 1990s. However, while Gro⁄ntoft has the most extensive chronology and has been thoroughly studied, more sites dating closer to the centuries around a.d. 1 provide further information on mobile village communities.
HODDE
Excavations at Hodde, Jutland, began in the 1970s. Hodde is typical of first-century b.c. rural settlement and has many traits that are present in Danish villages up to the beginning of the Viking Age. At its greatest extent, Hodde consisted of twenty-seven farmsteads. Each was composed of a longhouse with dwelling and cattle barn under the same roof, and a few smaller subsidiary buildings, perhaps barns or workshops. A fence surrounded each building complex, and a common fence, pierced by gateways affording direct access from each farmstead to its field, enclosed the entire village. There was an open area in the center of the settlement. One of the farmsteads, larger than the rest, may have been the residence of a chieftain. While some evidence of blacksmithing, pottery making, weaving, and spinning does exist, the primary economic activities were cattle breeding and crop raising, in keeping with the traditions of the Bronze Age but on a much larger scale. Other sites in Jutland show that, alongside such villages, there were also smaller agricultural settlements with only two or three farms, but we do not know why there were such great variations in the scale of settlement in the Danish countryside.
OTHER SETTLEMENTS
The evidence of house construction that is apparent in the Danish material cannot be detected in Köpinge, Scania (Sweden). Instead there are small, gradual changes. Continuity in settlement development in the Köpinge area—as in Denmark—from the pre-Roman Iron Age to the Early Roman Iron Age is apparent, in that many sites date to both periods. No stall partitions have been documented, unlike the case in Denmark. Nor does the relatively regular placement of the pairs of roof-bearing posts give us any guidance about the existence of stalls. Conversely, the length of the houses and the location of the hearths seem to indicate that one end was used as a barn, and that more animals were housed there than was the case in the Late Bronze Age. As with the structure of settlement, the archaeological material clearly demonstrates that these were isolated farmsteads. Only toward the end of the period do we find evidence of agglomerated settlements of two or more farms.
Information about the mobility of the Iron Age village society can also be gleaned elsewhere in Denmark, for example, in the low marsh regions by the North Sea in the south of Jutland, where the large migrating villages are characteristic of the period. At Drengsted, a very small area was found to contain a series of settlements, some with cemeteries dating from the first century b.c. to the fifth century a.d. At Dankirke in southwestern Jutland, a small area was found to contain several settlements, with their cemeteries dating from the same time period. In Gro⁄ntoft, Hodde, Drengsted, and Dankirke, the patterns seem to be identical. Over the centuries, mobile village communities centered around large herds of cattle moved around within narrowly defined resource territories.
REGIONAL VARIATIONS
The period of 200 b.c. to a.d. 200 is characterized by a warm, dry climate favorable for cereal cultivation. Descriptions of the cultural landscape and of land use in the Early Iron Age have long borne the stamp of the archaeological material from Jutland and areas preserving a fossilized cultivation landscape, such as Gotland and Östergötland. It is usually thought that most regions in southern Scandinavia underwent the same development at roughly the same time, not just of the cultural landscape but also in social and political terms; it is only as a result of differences in the form of the natural landscape in different regions that this development can be studied today, and then only in certain regions. In recent years large regional and local variations have become more evident, not just between areas with a fossilized cultivation landscape and those without, but also within each category. In Köpinge, it is impossible to know whether the farms in the area cooperated in any form of joint fencing, or what type of cultivation system was used. Analyses of carbonized plant material from the habitation sites show, however, that hulled barley had become the main crop and that weeds like Chenopodium had become more common, which indicates the presence of manured fields. The meadow plants in the material can be interpreted as hay waste and evidence of the stalling of animals. Traces of the production and working of iron have been documented. Iron extraction may also have led to the establishment of special habitation sites, as in the Krageholm area. Manuring and cultivation switch are also seen in the Bjaresjo area of Sweden.
CAIRN FIELDS AND CULTIVATION STYLES
In Norway in the early 1980s, there were systematic investigations of some cairn fields that had previously been interpreted as grave fields. Cairns are formed from clearing a field in preparation for plowing. They are simply rock piles. The typical clearance cairn field is characterized by a dense pattern of small cairns. These may belong to an extensive or to an intensive strategy of cultivation. The two strategies can coexist. Clearance cairn fields are characterized by a lack of internal boundaries, the usual evidence of a permanent arable field. The spatial organization of the cairn fields has no relation to the territorial division of farms from historical times. Phosphate analysis has located several settlements within the same cairn field. There are indications that the settlements had been abandoned and then used as arable fields. In some cases several phases of this cycle can be documented. This phenomenon is similar to what was occurring in the same period in the Danish village of Gro⁄ntoft.
Most prehistoric houses are found in Jaeren and Lista in Norway and belonged to the Roman Iron Age and the Migration period. They are three-aisled longhouses with stone walls. It was assumed that these house remains represented the first farms in Norway, which were the result of the climatic change in the pre-Roman Iron Age. That change forced people to house their cattle indoors and to collect winter fodder. Research in the 1990s and early 2000s indicates that the settlement change in southwestern Norway was caused by a shift to a more intensive type of cultivation. No one has found the houses from the first millennium b.c. because the farming system was based on bush fallow and shifting cultivation. In southwestern Norway, it was assumed that the clearance cairn field areas were evidence of extensive cultivation in the Bronze Age and the pre-Roman Iron Age. For a long time, however, settlement history in eastern Norway was written primarily on the basis of place names, graves, and archaeological artifacts. Extensive archaeological investigations in eastern Norway in the 1990s and early 2000s have located an increasing number of Bronze and Iron Age houses. More than twenty different settlement sites have been investigated, partly as research excavations, and partly in conjunction with rescue excavations (e.g., for the new Oslo airport at Gardermoen). These are found primarily in the presently cultivated lands—under the tilth. This means a large material culture is now available, consisting of buildings from the Bronze and Iron Ages.
The study of the principal house types that resulted from these excavations suggests that one principal type dominated from the Bronze Age to the Migration period. The three-aisled buildings were 15 meters long or more. As in contemporary cases from Denmark and Scania, there are indications of separate dwelling and cattle compartments. Each farmstead had two or three houses. This evidence dates the beginning of the "historical farm" to the Bronze Age. The cattle compartments show that cattle were stalled indoors and that winter fodder may have been collected. Within this system it must have been possible to collect manure and spread it on the fields. Therefore there is the possibility that an intensive type of cultivation was associated with the cairns.
The results of these investigations are consistent with the results from the cairns. Many house structures are contemporary with the field clearance cairns. The spatial organization of the cairn fields has no relation to the territorial division of the farms from historical times. At the site of Einang in Valdres, Norway, situated on the outlying lands of three different historical farm territories, the cairn field is located on the hillside, in an area which, in recent times, has been used chiefly as a pasture. The recent farmsteads, by contrast, are located along the valley bottom. They have prehistoric names and, in the graves associated with them, artifacts from the Late Iron Age have been found. In the clearance cairn, conversely, the graves contained artifacts from the Roman period. A pollen analysis shows that this area was cultivated continuously from the Late Bronze Age to the Migration period. The evidence from this locality points to a radical change in the structure of the landscape in the middle of the first millennium.
Sites from northern Norway show mixed economies of farming and fishing and individual farms rather than settlement complexes. Archaeological information coming from sites such as Bleik and Toften in Ando⁄ya point to a heavy exploitation of local marine resources and the beginnings of production of cured fish.
See alsoTollund Man (vol. 1, part 1); Hjortspring (vol.1, part 1); Emporia (vol. 2, part 7); Pre-Viking and Viking Age Norway (vol. 2, part 7); Pre-Viking and Viking Age Sweden (vol. 2, part 7); Pre-Viking and Viking Age Denmark (vol. 2, part 7).
bibliography
Batey, Colleen E., and James Graham-Cambell, eds. Cultural Atlas of the Viking World. London: BCA, 1994.
Becker, C. J. "Früheisenzeitliche Dörfer bei Gröntoft. West-jütland. 3. Vorbericht: Die Ausgrabungen 1967–68." Acta Archaeologica 42 (1971): 79–110.
Berglund, Björn E., ed. The Cultural Landscape during6000 Years in Southern Sweden: The Ystad Project. Ecological Bulletins, no. 41. Copenhagen, Denmark: Munksgaard International, 1991.
Callmer, J. "The Process of Village Formation." In The Cultural Landscape during 6000 Years in Southern Sweden: The Ystad Project. Edited by Björn E. Berglund, pp. 337–349. Ecological Bulletins, no. 41. Copenhagen, Denmark: Munksgaard International, 1991.
Derry, T. K. A History of Scandinavia: Norway, Sweden,Denmark, Finland, and Iceland. Minneapolis: University of Minnesota Press, 1979.
Fabech, Charlotte, and Jytte Ringtved, eds. Settlement andLandscape: Proceedings of a Conference in Århus, Denmark, May 4–7, 1998. Moesgård, Højbjerg, Denmark: Jutland Archaeological Society, 1999.
Hedeager, Lotte. Iron-Age Societies: From Tribe to State inNorthern Europe, 500b.c. toa.d. 700. Cambridge, Mass.: Blackwell, 1992.
Helliksen, Wenche. Gård og utmark på Romerike 1100
f.Kr.–1400 e.Kr.: Gardermoprosjektet [Farm and periphery in Romerike from 1100 b.c.–a.d. 1400: The Gardermoen project]. Varia, no. 45. Oslo, Norway: Universitets Oldsaksamling, 1997.
Hvass, Steen Hodde. Et vestjysk landsbysamfund fra ældre jernalder [A western rural society from the Older Iron Age]. Arkæologiske studier, vol. 7. Copenhagen, Denmark: Akademisk forlag, 1985.
Jensen, Jo⁄rgen. The Prehistory of Denmark. New York: Methuen, 1982.
Myhre, Bjo⁄rn. "Agrarian Development, Settlement History, and Social Organization in Southwest Norway in the Iron Age." In New Directions in Scandinavian Archaeology. Edited by Kristian Kristiansen and Carsten Paludan-Müller, pp. 224–271. Copenhagen: National Museum of Denmark, 1979.
Näsman, Ulf. "Hus, landsby, bebyggelse [House, rural settlement, and structures]." In Danmarks længste udgravning: Arkæologi på naturgassens vej 1979–86 [Denmark's longest excavation], pp. 69–86. Copenhagen, Denmark: n.p., 1987.
Sophia Perdikaris