Skip to content
IUScholarWorks Journals
23.10.02 Sørensen, Gudme: Iron Age Settlement and Central Halls

23.10.02 Sørensen, Gudme: Iron Age Settlement and Central Halls

Gudme, a small municipality on Danish island Funen, has been of interest since the discovery in 1833 of the Broholm Treasure, gold jewelry from the Iron Age. Striking finds had raised interest in the area before, in 1993, a “great hall” was found there during preliminary digging for a local sports stadium. Two comparable halls recently excavated at Lejre on the Danish island Zealand, named in Ynglinga Saga and other sources, were both built later. [1] So the Gudme area, where more gold from the Danish Iron Age has been discovered than anywhere else, is also the site of the earliest Scandinavian “great hall.” With about fifty farms in a period when a good-sized village had twelve, Gudme was a large settlement. Over 6,000 metal objects dating from the period 200-1100 CE have been recovered, including at the coastal landing site Lundeborg several kilometers away. Finds from the fourth through sixth centuries include six gold hoards and other gold objects with a total weight of more than ten kilos (twenty-two pounds), as well as five silver hoards.

Written by an archaeologist who took part in the 1993 discovery--now a curator of Roskilde Museum--this volume is a welcome “first comprehensive presentation” of the locale, widely regarded as one of the earliest “central places” of the Scandinavian Iron Age (Jessen, “Preface,” ix). [2] Such sites raise the issue of state formation, with religious authority and political power linked to craft production in precious materials and to supra-regional trade. Sørensen does not define “hall,” but associates the term with (1) monumental size and location; as well as (2) high-status goods, that (a) would not be found in an ordinary farm building, or (b) were used to invoke supernatural powers. He names as halls both the large Gudme building where numbers of people could gather, and most phases of the smaller one for restricted-access rituals. The large structure, “great hall,” he sometimes calls a “special banqueting building,” the smaller a “cult building or temple.” [3] A cult building was erected first, and became a “hall” itself in Sørensen’s Phase 3, when, with comparable architecture, it accompanied the “great hall,” which “functioned for a period of up to 140 years, from c. AD 275 until around AD 420” (29).

Sørensen analyzes hundreds of postholes, wall trenches, fence lines and other construction remains. Using stratigraphy, datable finds, and radio-carbon assessments, he has created a 650-year chronology of about a dozen building sites in central Gudme. Seven phases of the temple building overlapped with two phases of the great hall before they were succeeded by two phases of farm sites. To the west separated by wetlands, three farms that worked precious metals, like most near the halls, went through nine iterations. To the east two other workshop farms developed through twelve phases.

The material record shows significant religious innovation in Gudme. At Uppåkra, Sweden and probably at Sorte Muld, Bornholm cult buildings were erected at roughly the same time as at Gudme (112-17), but no great hall is known at either location. Gudme is unique in having had a cult building at an early point in northern European history closely associated with a great hall. The Gudme great hall was indeed monumental: 47m long (154ft) and, at its widest, 10m (33ft), with roof-bearing posts 80cm (33in) in diameter. Most “great halls” are later than the one at Gudme, although Norway is now reporting Roman Iron Age farmhouses that had special-purpose rooms (118, 105-107). Religious customs were reshaped to begin the site, and it clearly became an area of political and religious conflict.


Sørensen notes that Tacitus’s Germania was written about seventy-five years before the earliest cult building at Gudme. The closeness in time poses a conundrum, because Tacitus “describes how, certainly among some of the Germans, the cult is practiced in the open air and...buildings were not associated with the cult site [nec cohibere parietibus deos...arbitrantur (Germania, Chap. 9)]. This contrasts with the situation at the Gudme halls, where the buildings are the most concrete indication of the cult practice that must have been undertaken at the site” (91-92).

Sørensen opines that: the “dedicated cult building should be seen as...inspired by the Roman area along the Rhine, where during the first and second centuries [CE] the Germanic peoples came into contact with the Romans and their cult practices, which were associated with buildings in the form of temples” (92).

Further changes were to come. In a pit probably associated with the end of the first two phases of the cult building (about 275 CE), the remains of a ceremonial carriage were found. Sørensen (19, 92) considers it comparable to the carriage (vehiculum) used in a festival described by Tacitus (Germania,40) in which seven Germanic tribes, including the Angles who would give their name to England, venerated Nerthus, Mother Earth (Terra Mater). Sørensen links some of the Gudme carriage parts with those recovered at Dankirke, Jutland, where buildings to house a carriage may have stood through the first half-millennium of our era. [4]


In the fifth century, instead of a rebuilding of the great hall, “a much smaller hall building was...erected 60m to the north-east” in a significant shift. “The close connection between the secular domain of the ruler in the form of the large hall building and the cult site...appears to have been severed at the beginning of the 5th century,” concludes Sørensen (96-97). “Even if house XV can still be interpreted as the ruler’s banqueting hall, the balance between the two buildings, and thus between the two institutions, has by now tipped in favor of the cult building...This change...possibly reflects the old ruling family, who had erected the great hall, coming under pressure from other families or groups within the area, and therefore being unable to maintain control over the cult activities that took place at the site. Another possibility is that there may have been a dispute between the ruler and the priesthood, which caused the former’s power over the cult site to diminish. There must, under all circumstances, have been radical changes, which were further accentuated during the next phase of the settlement.” [5]

Both these possibilities could have been in play in a religious shift, with the ruling family losing control of cult activities that developed in a different direction. Two sites besides Uhrenholdtgård, each with over 300g (0.66 lb) of gold, identify “an area of high ranking warriors” separate from the Gudme halls and workshops (98; Jørgensen 1995a).

After the “great hall” was relocated and reduced, “a small silver mask depicting a male face” (Fig. 1.68) was dug down into a posthole of the rebuilt cult building, “perhaps as an offering associated with the [rebuilding].” Sørensen considers the mask “a depiction of one of the male deities of early Nordic mythology,” and speculates that “much larger, carved cult figures made of wood… may have stood in the cult building itself” (97). If so, this may have been an innovation.

When the cult building was rebuilt the last time a half century later, the smaller “great” hall (House XV), “based on the available evidence,” was not rebuilt at all. “The aristocratic presence at the cult site, in the form of the secular banqueting hall, is now gone, and the development that had which the more independent cult site was freed from the ruler’s domain, had apparently been completed” (97).

A gilded silver “coin” associated with the end of the cult building is “totally unique” in Scandinavia and Europe, but related to a number of developments. Its two interlaced Style I animals are matched on scabbard ornaments of gold found nearby, including one weighing 99g (nearly 1/4lb), the finest known from Denmark, which came from the Uhrenholdt farm, with three hoards, two gold, one silver, about a kilometer from the hall site. Sørensen points to D-bracteates as “another finds group” with somewhat similar design, particularly bracteates from Sletner and Kydland, Norway (70). “The small ‘coin’ reflects the end of the hall settlement, which ceases to exist around (the middle of the 6th century)” (71).

“Sherds from a stamp-decorated with Anglo-Saxon England” (98) in the last phase of the cult building. A similar straw-tempered ceramic from the roughly contemporary oldest great hall at Lejre, is likewise seen as an import from early medieval England (85). The sites where the Gudme great hall had stood until about 420, and its smaller version until about 470, were absorbed into a new farm whose main building shared architectural features with the halls that were raised in the sixth century at Lejre and Tissø.

Gudme on Funen

The name Gudme, recorded in the thirteenth century for the Funen site, means “home of the gods” <Old Danish guth + hēm. Other sacral place-names, such as Albjerg, “Hill of the Shrine,” Offerbjerg, “Hill of Sacrifice” occur nearby. It has been suggested that place-names show the Gudme landscape to be a “paradigmatic model of Asgard,” abode of Norse gods the Æsir. [6] This view has received some critical support. However, ten other sites in Scandinavia, including four more in Denmark, bear a version of the name Gudme. This“Gudme/Gudhem Phenomenon” was the subject of an interdisciplinary workshop in Schleswig in 2010 which considered onomastic, historical, archaeological, and topographical aspects, without finding common characteristics among the eleven sites. [7] Nonetheless, one “working hypothesis” for the Phenomenon was that “the place names originally referred to lakes, rivers, bogs or...[an]other wetland environment as the ‘home of the gods’ and that the settlement names Gudum, Gudhjem or Gudme were secondary developments.” [8] Possibly related, the name for the largest Danish river, Guden(å) <Guthn or Guthano signifies “dedicated to the gods.”

Explaining the hypothesis, Andres Dobat noted that “watery or wetland environments” have been a “consistent element in a number of sites with religious/cultic functions,” and pointed to texts about the Norns, the gods’ Thing councils, and the abodes of Frigg, Saga, and Iðun. “The idea of divine beings being closely related to water emerges as a well-established concept in pre-Christian cosmology...Seen from this perspective, not the settlement at Gudme, but the lake at Gudme would have been referred to as the ‘home of the gods.’ Taking the argument even further, the natural feature of the lake at Gudme can be seen as the primary element, with the settlement at Gudme only a secondary element, which drew its legitimacy as a religious central place from the sacred properties of the lake.” [9]

Sørensen’s account supports this hypothesis on Funen, and is worth quoting at length: “a final important element related to the emergence of the cult site at Gudme was probably Gudme Lake, located 600 m west of the settlement. The very deep lake fed by upwellings, with its special location in a marked hollow in the landscape, is associated with a mystical light. The incorporation of large parts of the landscape into the cultic area is probably also reflected by names, which which in some way must have been involved in the cult activities centered at the Gudme halls. It is...worth noting that the areas immediately south and east of Gudme Lake were never part of the inhabited area. Between the westernmost part of the Iron Age settlement and the lake is a 250-m-wide, uninhabited area, which topographically would otherwise have been very well suited for settlement. The higher areas alongside the lake are actually the only places from where it would have been possible to overview its surface. It is tempting to assume that these areas were kept free from settlement due to cultic activities that took place near the lake. But there is still no concrete evidence which indicates that the lake was a sacrificial site. Excavations of the lake have demonstrated that it contains a very thick layer of sediment, so that it was not possible to reach the level of the Iron Age lake bottom” (93).

Sørensen’s careful recording of buildings has allowed him to formulate a typological principle for dating northern European Iron Age houses. Those with roof-bearing posts having cross-spans in the end walls at least 0.4m (1.3ft) shorter than those at the house’s widest point are considered to have a curved line of posts--an innovation that begins in the first half of the sixth century (344). This criterion makes it possible to date houses without chronologically diagnostic finds.

Much of Sørensen’s presentation reads as an excavation report. However, the equivalent of “Discussion” sections (1.6, 2.9, and 3.7) provide helpful narrative summaries at the ends of chapters. Cross-referencing sometimes has broken down. For example, subdivisions of Section 1.3 do not exist, although they are referred to at least 15 times (12, 15, 18, 19, etc.). The lack of an index diminishes the value of this very detailed presentation--this review has sought to present some of the key findings that might be obscured for the more general reader. In the treatment of Lejre “(Fig. 1.106, House II and III)” is mistakenly discussed (102): “In each corner of the younger building (House II)...” should read “In each corner of the older building (House II)...,” because House II of Fig. 1.106B is older than house III (label erased in the Figure) of Christensen’s graphic for Fredshøj (Fig. 1.106B), and older than House III of Christensen’s graphic for Mysselhøjgård (Fig. 1.106A).

These copyediting errors aside, Sørensen has fastidiously and valuably constructed a data set about more than half a millennium of human activity in a focal site where political and religious changes that we will continue to probe shaped the fabric of northern Europe.



1. Tom Christensen, “Lejre Beyond Legend--The Archaeological Evidence,” Journal of Danish Archaeology, 1991, 10: 163-85; Lejre Bag Myten, De Arkæologiske udgravninger (Moesgård: Jysk Arkæologisk Selskab, 2015); John D. Niles, et al., Beowulf and Lejre, (Tempe, AZ: ACMRS, 2007).

2. Frank Battaglia, “Not Christianity versus Paganism, But Hall versus Bog: The Great Shift in Early Scandinavian Religion and its Implications for Beowulf,” Anglo-Saxons and the North, ed. Matti Kilpiö, et al. (Tempe, AZ: ACMRS, 2009),47-67; Karen Høilund Nielsen, “Key issues concerning “central places,” Wealth and Complexity, ed. Ernest Stidsing, et al.(Aarhus: East Jutland Museum, 2014), 11-50.

3. “Private” banquet halls may have existed (118), but not at Uhrenholdtgård (98).

4. Morten Søvsø, Ribe 700--1050, From Emporium to Civitas in Southern Scandinavia (Moesgård: Jutland Archaeological Society, 2020),103-10.

5. Sørensen conceives Beowulf as “associated with Lejre” (102), but this conflict is surely represented in it.

6. Lotte Hedeager, “Asgard Reconstructed? Gudme--a ‘Central Place’ in the North,” Topographies of Power in the Early Middle Ages, ed. Mayke De Jong, et al. (Leiden: Brill, 2001), 467-507.

7. Oliver Grimm and Alexandra Pesch, ed. The Gudme/Gudhem Pnenomenon (Neumünster: Wachholtz, 2011).

8. Andres S. Dobat, “What was the ‘home of the gods’? Gudum in Flask Hundred and Gudum in Skodborg Hundred in northern Jutland, Denmark.” In Grimm and Pesch, 91-110, at 91-92.

9. Dobat 107-108.