Håkansson, Beyond the Archetypes. Social Hierarchies in Rural Medieval Halland ??–??

Big bang, lordship or inheritance? Changes in the settlement structure on the threshold of the Merovingian Period, South-Eastern

Crisis, Herrschaft oder Vererbung? Veränderungen in der Siedlungsstruktur an der Schwelle der Merowingerzeit, Süd-Ost-Norwegen

Crise, la seigneurie ou d'un héritage? Changements dans la structure de règlement sur le seuil de l'époque mérovingienne, du Sud-Est de la Norvège

Frode Iversen

Introduction

Between 500–700 AD, settlement structure in Scan- explain these changes have been related to settlements dinavia changed radically. More and more abandoned stagnation, a decrease in population or plagues. In settlements have been discovered, including in the contrast to this view, 75 per cent of the now known central arable farmlands where they were previously sites in the south-eastern region of Norway have been unknown. It has not been fully understood what has investigated since 1990, which makes me ask: can the caused this large-scale abandonment, but several hy- image of settlement crisis and stagnation be main- potheses have been proposed. In this paper, I examine tained when considered using the new evidence from these hypotheses more closely, drawing upon archaeo- recent excavations? logical evidence from settlement sites in eastern Nor- New research in Scandinavia has provided valuable way that have not been included and studied in this knowledge on the spatial organisation of households, line of discussion so far. The more traditional ‘crisis farmsteads and agrarian landscapes (e.g. Ahlkvist hypothesis’ is evaluated against what I have termed 2002; Edblom 2004; Streiffert 2005; Söderberg 2005), as the ‘lordship’ and the ‘inheritance’ hypotheses, ap- as well as on the relations between sites and settle- proached both from a macro and micro perspective. ments in a wider perspective (e.g. Fabech – Ringtved In the administrative region of the Cultural Museum 1999; Carlie 1999; Göthberg 2000; Helgesson 2002). It of History in Oslo (CHM), a total of 139 sites produc- is generally agreed upon by Scandinavian archaeolo- ing evidence of 450 houses have been excavated as of gists that the settlement pattern went through major 2010 (Fig. 1). Almost half are located in the lowlands, changes on the threshold of the Meronvingian Period. the co-called claylands of the counties Østfold, Vest- The reasons, however, are contested. fold and Akershus, and not the least the Ra in Østfold, Romerike and in the low-lying rural areas close to the Three hypotheses Ra in (Bårdseth 2007; Gjerpe 2008; Martens 2007; Mjærum – Gjerpe 2012a; 2012b). Only a few exca- The ‘crisis hypothesis’ has been further developed vations have been carried out in the valleys, woodlands and elaborated in recent research by Bo Gräslund and the mountain areas in the south-eastern part of (2007). In this context he has looked more closely at the Norway because of less development pressure. phenomena Fimbul and Ragnarok referred to in mythi- Before 1990, most Scandinavian research on pre- cal accounts and verse, such as Gylfaginning and Ka- historic settlements and farmscapes were based on leva. He also fi nds arguments using written sources fossilised house-foundations and fi eld systems, as from Late Antiquity and China for placing the start of they are usually deserted and preserved in marginal a climatic crisis to 536AD and lasting several years, environments, especially in the southern part of Nor- which is also supported by scientifi c evidence. Between way. (e.g. Myhre 1972; Pedersen 1990) (Fig. 2). A widely March 536AD and September 537AD, two summers al- held explanation has been that the abandoned farm- legedly were ‘lost’, which resulted in a long continuous steads represented a periphery phenomenon caused cold spell (Gräslund 2007, 104 f.). This view is support- by a contraction of settlements. The theories used to ed by abnormally high quantities of sulphates found

?????????????????, ed. by Jan Klápště & Petr Sommer, Ruralia, IX (Turnhout: Brepols, 2012), Pp. ??–??. FHG 10.1484/M.RURALIA-EB.1.100158 Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig.1. Excavated settlement sites in South-Eastern Norway per 2010. N=139. Data from MCHs settlement research group (see acknowledgement). Map: Frode Iversen.

2 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

in the East Antarctic ice cap (540+/–17 years) and in 2011, 144). Elites have come and gone through history. Greenland (534+/–2 years), which has been interpreted One might consider a model based on a reorganisa- as traces caused by a volcanic eruption or an asteroid tion of settlements in central areas and lordship-driven impact. Dendrochronological evidence from Europe, clearances in the periphery during the Merovingian Asia and South America shows diminished growth in and Viking Periods. the years after 536AD. This suggests that the crisis Studies concerning settlement layout surrounding seems to have had a profound global impact, and may the high-status site of Gamla Uppsala have shown a signify a temperature drop of 4–5 degrees Celsius in signifi cant change around 600 AD (Göthberg 2007, the Northern hemisphere of Scandinavia. Recent stud- 442). Similar changes have been observed in Östergöt- ies of European climatic development also support this land near Linköping and on sites in Småland, South view (Büngten et al. 2011; Zachrisson 2011, 145). Analy- (Ericsson 2001; Petersson 2011, 251). While sis of pollen samples suggest an growth of vegetation farmsteads were relocated, the arable fi elds were con- in the cultural landscape and increased forestation in tinuously used (Petersson 2006, 32). These fi nds sub- the Middle- and Northern Europe during the last dec- stantiate a model that implies that minor farms were th ades of the 6 century. abandoned for the prospect and establishment of larg- The hypothesis Gräslund presents is fascinating. er nucleated settlements (Myhre 2002; Zachrisson 2011, One single catastrophic event sets off a global crisis 144). In British research, the so-called manoralisation – one factor, a big bang. The crisis might have been and development of open-fi eld systems, has also been enhanced by the Justinian plague and thus created a main topic of discussion (Reynolds 1999). extreme conditions. The results were population cri- Rich monumental burials have also been taken as in- sis and destruction en masse in liminal areas of agri- dicators of a concentration of power and the emergence cultural land in the North. The impact was especially of new elites in Eastern Norway during the Meroving- profound in settlements in colder climatic zones, such ian Period (Myhre 1992; Gansum 1995). Rich graves in as Scandinavia. Gräslund links the abandonment of the so-called Åker complex also emerged around 600 farms during the Iron Age to this crisis (Fig. 2), but AD (cf. Martens, this volume). Terje Gansum (1995), does not provide any in-depth analysis of when these who has examined the large mounds in Vestfold, found settlements were abandoned or how they relate to the that although there is approximately the same number surrounding agricultural landscape. The available evi- of large mounds dated prior to the 600 AD as after, the dence does not indicate a long-lasting climatic crisis. earliest are more scattered in the landscape. The later The question is, however, if this climatic crisis did mounds are generally bigger and often concentrated have an impact on the settlement pattern to such ex- around archaeologically documented power centres, tent as Gräslund presumes. More accurate dating with such as Borre, Sem and Gokstad, and indicate nucle- regard to the time of the abandonment to the decades ated power during the Late Iron Age.Similar perspec- after 536AD would perhaps strengthen Gräslund’s hy- tives have been discussed for other regions in Norway pothesis. This in turn would present a foundation for and Sweden, using large mounds and rich graves as a discussion concerning the extent of the crisis and its points of reference (cf. Iversen 2008, 76; Bratt 2008). It consequences for settlement layout. remains to be seen, then, if the abandoned settlements The so-called ‘lordship’ hypothesis that started to can be connected to the establishment of large estates gain support in the 1990s, is based on a process-ori- surrounding such power centres. ented model that ties landscape and settlement organi- A third, and less discussed hypothesis emphasises sation to the development of petty kingdoms and elites changes in hereditary rights from primogeniture to holding larger estates. Previously, it was argued that partible inheritance. Torun Zachrisson (1994; 2012) has Germanic prehistoric societies were almost exclusively discussed the impact of allodial rights in relation to the made up of free farmers and families of equal standing. organisation of Iron Age communities. A ‘fi ve-genera- The so-called Gemeinfreie-thesis is now outdated. Today, tion rule’ seems to characterise land inheritance dur- this society is considered as hierarchically organised, ing Late Iron Age, according to runic inscriptions from and with large estate owners (Rösener 1995; T. Iversen Scandinavia (Brink 2002, 103ff). It has been pointed 1997; Skre 1998; F. Iversen 2008; 2009). The origin and out that the last stem in the older runic alphabet the the development of systems of landownership in Scan- futhark is called ‘odel’, which has its origins from Ro- dinavia has, however, been poorly understood. man times (Robberstad 1967, 493, Erikson – Strid 1991, The question is whether these abandoned settle- 24; Zachrisson 1994, 219). In a system with partible in- ments are connected to the emergence of new elites who heritance large estates would slowly be divided among reorganised the landscape, as Dagfi nn Skre (1998) and multiple descendants, while primogeniture would have Bjørn Myhre (2002) have proposed for Norway. Similar the effect of keeping large estates united by the control models have been discussed for Sweden (Zachrisson of the elite.

RURALIA IX 3 Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 2. Bo Gräslund has seen the many abandoned farms in Scandinavia in the context of a comprehensive climate crisis and the volcanic eruption in 536 AD. The farm Hanaland in Time Jæren was abandoned in the Migration period and taken up again before year 1000, before it was fi nally deserted in the late Middle Ages. Photo: Myhre 2002, drawing: Myhre 1972.

We do not know much about hereditary rights in pothesis concerning a shift in hereditary laws can be Scandinavia during Early Iron Age but Tacitus offers linked to the ‘lordship’ hypothesis. The kings’ follow- some insight concerning the Germanic areas in 98 AD ers may have gained more importance with the emer- (Ch. 20, 32). He seems to assume primogeniture as gence of new petty kingdoms. Warriors who did not the main rule of succession amongst the people called own land themselves may have achieved better con- Tencteri (Germ, ch. 32), located near the Rhine, close ditions through their expanding military and politi- to today Köln and Bonn. By law, slaves, house and cal role. The question if farmsteads and their houses land was inherited by the eldest son. But the horse, ac- were abandoned as a consequence of the division of cording to Tacitus, was inherited by the son who was infi elds and the establishment of new farmsteads, still considered the best warrior, regardless of age. By em- remains without an answer. This issue represents two phasising this as an exception to the rule, Tacitus illus- contradicting explanations. On one hand, we are faced trates the strong warrior ideal amongst the Tencteri. with the possibility of depopulation in a time of crisis. There are, however, methodological problems in- On the other hand, there is also signs of population volved by using Tacitus as an analogy for early heredi- increase and intensifi cation of land use, including di- tary laws in Scandinavia. Primogeniture is not known vision of farms. A central issue is related to time. in so-called provincial laws, recorded in the 1100s and In the following I give a brief overview of excavated onwards. The main principle was partible inheritance settlement sites in south-eastern Norway. The aim is with a gradual system of defi ned classes of heirs. As a to identify certain traits in the material that can illu- general rule land was inherited equally among sons, minate the hypotheses and also to demonstrate the and movables were split equally among daughters. Ac- potential the archaeological material has to recognise cording to the Gulating law of Western Norway, daugh- changes in land use and peasant society. ters could also inherit peripheral farmland in the out- skirts (Gjerdåker 2001, 15; Kleven 2001, 64–71). The Methods and material law has certain and distinct stipulations concerning the division of land. If no other methods were found, The large-scale excavation of settlements at For- the infi elds could physically be divided in two halves, a sandmoen in Rogaland 1984–1993 was important in solution that easily would have caused separate settle- many respects, not least that a new generation of Nor- ment layouts without shared ownership of the infi elds wegian archaeologists developed competence in meth- (Stylegar 2005). ods concerning top soil stripping (Løken – Pilø – Hem- So far, I have presented one event-oriented hypoth- dorff 1996; Løken 2001). A new and more effi cient way esis and two process-oriented hypotheses. The hy- of studying agricultural settlement in the lowlands of

4 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Norway was introduced, and adopted for use in the that pole-built houses had a life expectancy of 25–50 many rescue excavations that increased in number in years, but under optimal conditions considerably long- the 90s and the new millennium. Every year between er. Material from south-eastern Norway suggests, how- 100 and 150 excavations are carried out in Norway. ever, that the houses seldom can be dated more exact- About 50–75 per cent of these are excavated by topsoil ly than within a timeframe of 100–200 years (Gjerpe stripping and on sites that usually include traces of – Østmo 2008, 134). This makes it diffi cult to establish settlements. a fi ner chronological time sequence with regard to the The Museum of Cultural History of Oslo is the larg- abandonment of farmsteads. est excavator in Norway. Its district covers ten out of The sites within this study are dated to the following 19 counties and 55 per cent of the farms in Norway periods; Roman Age, Migration Period, Merovingian (30,880 of 55,688 farms, according to the survey of Period and the Viking Age. The absence of Meroving- Norske Gaardsnavne). By 2010, excavations of prehis- ian Period houses in already established farmsteads is toric house sites have been carried out at 132 farms, taken as an indicator of abandonment. The three-aisle which represent 0.4 per cent of the total number of longhouse with the division of living space and an area farms, including 139 sites with traces of 450 houses. for animal husbandry is a common feature through- The ratio of houses to sites varies from 1 to 25. My out the Iron Age, but becomes gradually less common study here concerns the settlements at nucleus of during the Viking Age (Martens 2007, 102f). In the farms (Norw. tun) that have been in use through gen- Museum’s administrative region, only nine three-aisle erations before they were abandoned. I therefore dis- longhouses have been C-14 dated to the Merovingian tinguish between tun and a house plot. A tun is defi ned Period, as opposed to fi fteen to the Migration Period by as a settlement that comprises several house plots, 2002 (ibid.). preferably three or more. Abandoned houses are also Hypothesis 2 emphasises the social context and his- found in the outskirts of larger historically known tun torical known settlement structure. So far, the excavat- (i.e. Hesby, Vestfold) and are, as such, more uncertain ed material has been discussed only to a limited extent as indicators of abandoned farms. and incorporated in a ‘social landscape’ perspective, About 64 per cent of the material, 89 out of 139 incorporating the methods of historical archaeology. In sites include traces of only one or two houses (roughly Norway such methods have especially been developed 60 and 29 sites). Their relevance as a source for this since 1995 at the University of Bergen in the research study is therefore limited. Twenty-two sites, about 16 environment around Professor Ingvild Øye. The stud- per cent, have traces of three or four houses. I conse- ies combine interdisciplinary methods, using written, quently regard these as more representative. The re- archaeological and cultural geographical evidence in maining 20 per cent, 28 sites, have traces of fi ve or a long-term perspective that includes and emphasises more house plots. Fifteen of the sites have fi ve or six the whole agricultural landscape and the social and eco- houses and 13 sites have traces of seven or more house nomical relations between the settlements (Øye 2002a; plots, and can be recognised as farmsteads that were 2002b; Iversen 2008). The methods have only to a lim- abandoned. These should therefore have a potential of ited extent been applied in rescue archaeology, that of- shedding light on the processes behind the abandon- ten focuses on the technical and architectural aspects ment, although the archaeological material as such is of the houses and less on their context of the historical relatively modest. landscape and farm land (e.g. Diinhoff 2011). Different methodological steps have been taken Position is a key feature when exploring hypothesis to approach the three hypotheses. The time of aban- 3 related to possible hereditary divisions. Here, I have donment is a key element of the fi rst hypothesis. It is, assessed the sites by taking into account so-called however, diffi cult to determine the exact time the last division names, which represents another new meth- house in the tun went out of use. Jes Martens have dis- odological approach. Out of 56,688 farms recorded in cussed such problems thoroughly in the case of south- Oluf Rygh’s voluminous compilation, Norske Gaard- ern Kjølberg in Østfold, where four longhouses may navne (1897–1936), altogether 7150 farms have prefi x- represent four generations at the same farm (Martens es, such as northern, southern, eastern, western, big, 2007, 96–98, 104). The dating is based on the latest C- little, upper, lower etc., and thus an indicator of divided 14 date of constructive features, such as roof-bearing farms. An abandoned farmstead situated by a bound- poles, by which the age of the wood itself is measured. ary, e.g. between two historic farms carrying division Alternatively, dating can be provided by secondary de- names, may also indicate that such a process have posited material with a relatively short time horizon, taken place. In earlier research, farm-names have been e.g. a single charred grain. Such dates do not always considered as a reliable source for understanding the correspond, even when they are taken within the same development of prehistoric farms. Together with Mag- house plot. Haio Zimmermann (1998, 60–62) suggests nus Olsen’s work Ættegård og Helligdom (1926), Rygh’s

RURALIA IX 5 Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 3. Percent of Farms with division-names per county in Norway: N = 7150 of 56,688. Based on data from . The data from South and North Trøndelag are not polite.

work revolutionised place-name research and laid the suggest that the physical farm division was more com- foundation for later settlement studies. With addition mon in central areas of Eastern Norway than in other to chronological layers, the toponymic material also parts of Norway (Lunden 1969, 93; Stylegar 2005). refl ected different social levels. Thus it was a useful I have chosen two sites in Vestfold for closer exami- source, but also had its limitations. New archaeologi- nation to assess the validity of the hypotheses with an cal information about prehistoric settlement offers, emphasis on the relation between the sites and the his- however, new and more direct ways of assessing the torically known settlement patterns. How were these same material. settlements situated in the landscape when compared It is unclear how far back it is possible to trace di- to the historical farms, the organization of the infi elds vision names. They appear in written sources around and outfi elds in the Middle Ages and farm bounda- the 13th century. In the eastern part of Norway, Bishop ries? Both sites have been excavated as part of the E18 Øystein’s cadastre, the so-called Red Book, lists such road project led by the archaeologist, Lars Erik Gjerpe. names systematically for church property from the Site 1 (Auli) was excavated in 2009–10 by Line Grind- 1390s. The Red Book also refers to abandoned medi- kåsa, and site 2 (vestre Ringdal) by Marit Østmo in eval farms with division names. A study on prehistoric 2005. The excavation results are published by Gjerpe graves located close to farm boundaries in Vestfold, (2008) and Mjærum – Gjerpe (2012a; 2012b). includes farms with division names, and may indicate Site 1 Auli is located close to the historically known such processes in the Iron Age (Ødegaard 2007; 2010). royal villa of Sem. The farm belonged to the St Olav’s The question then is if these processes could be monastery in Tønsberg before 1390. Several scholars, connected to elites who were reorganising the land- among them Halldis Hobæk (2008), have convincingly scape and emergence of new hereditary laws, and be argued that Auli prior to the foundation of this mon- recognized as indicators of intensifi cation and popu- astery in 12th century was part of a large royal estate lation increase. Such names occur most frequently in consisting of nearly fi fty subordinated farms under the counties Vestfold and Akershus that also happen the manor of Sem. The site is therefore located in a to be the areas that have the highest number of aban- distinct ‘lordship’ environment. Site 2 is located in an doned farmsteads. More than 30 per cent of the farms area that had many freeholders during the late Middle in Vestfold, 24 per cent in Østfold and 22 per cent in Age (Ødegaard 2007, 73; 2010). The reason why these Akershus have names that imply divisions (Fig. 3). The two farmsteads were abandoned will be assessed and percentage of division names decreases the further discussed against the main tendencies in the source west and north of these regions one looks. This may material.

6 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Results These fi gures are in contrast to the development in the transition phase to the Merovingian Period with Fig. 4 shows the average number of abandoned very slight signs of continuity. Only three of nineteen sites per century and reveals an interesting pattern. It settlements that have traces of fi ve or more houses, demonstrates that there is a four to six times greater and settled in the Migration Period, show later activity. frequency of farm abandonment during the Migration This is especially true for Moi in Vest-Agder with con- Period than in the Iron Age as a whole. When review- tinuity through the Middle Ages (Reitan 2009). This is ing the Iron Age and Middle Age, there is a pattern of also the case for southern Kjølberg in Østfold (Martens three (Merovingian Period) and up to fi ve and six (Mid- 2007), but more uncertain at the site vestre Ringdal in dle Ages) abandoned settlements each century. This Vestfold (Gjerpe – Østmo 2008). The fi ve longhouses number also includes small sites. The Migration Pe- at southern Kjølberg are located close to the present riod stands out, with a total number of 19 sites per farm. Finds dated to the Viking Age and Middle Ages century (38 out of 139 sites). This is further enhanced might suggest that the original settlements have been by a closer look at the sites that contain traces of fi ve moved to its present location (Martens 2007, 96). The houses or more. Approximately 16–17 of 28 sites that longhouse that signifi es the latest phase at western have this attribute were abandoned during the Migra- Ringdal in Vestfold (House II) has been C14-dated to tion Period. Five were abandoned during the Middle Early Merovingian Period or Late Migration Period Ages, four in Pre-Roman Iron Age and two to three (Gjerpe – Østmo 2008, 57). Charred grain, considered were abandoned during the Merovingian Period. Trac- a secondary deposition in two roof-bearing postholes, es of larger abandoned settlements dated to the Roman has been used to date the house. The remaining twelve Period and Viking Age have not yet been uncovered. longhouses at western Ringdal have been dated to the Six settlements from the Migration Age are located in Roman Period and Migration Period, and the settle- Akershus and four in Østfold. Similar types have been ments seem to have terminated by the end of the Mi- found in the county of Vest-Agder, two in Vestfold and gration Period, c. 600 AD. Thus, it is only Moi in Vest- one in each of the counties of Buskerud, Hedmark and Agder that can prove continuity in settlement with . any certainty through the Merovingian Period. This As for settlement continuity, approximately 40 out suggests that many settlements were abandoned dur- of 56 sites (71 per cent) have houses dated both to ing the Migration Period in Eastern Norway. A similar the Roman Period and the Migration Period, including pattern has been identifi ed in other parts of Norway, small sites. However, large settlements show continu- based on fossilised houses in marginal areas. Now this ity through the Roman Period to Migration Period. A trend has also been recognised in central agricultural total of 24 out of 27 settlements (89 per cent) from the areas in Eastern Norway as well. Roman Period (with three or more houses) were settled But what about sites at farms with division names? also in the Migration Period, and 19 out of 20 settle- Altogether 58 out of 139 sites (42 per cent) are located ments having fi ve houses or more (95 per cent). at such farms. In Østfold this is the case for 20 out

Fig. 4. N = 139. Abandoned sites per century (2400 BC–1500 AD), South- Eastern Norway. Data, see acknowl- edgment.

RURALIA IX 7 Frode Iversen, Big bang, lordship or inheritance? ??–??

of 31 sites, in Vestfold 11 out of 23 (including Nordby), names. This phenomenon is, however, not limited to and in Akershus 15 out of 30 sites. In total, 46 out of the Migration Period. To explore this relation further, 84 sites (55 per cent) are located at such farms in these two case studies of sites located at such farms will be three counties. This seems to be a recurrent feature. considered and included in the fi nal discussion. The regular frequency of such names is 26 per cent (2258 of 8795 farms). In the other counties in East- ern Norway, the numbers are too small to draw parallel Results from the case studies assumptions. Closer assessment of sites dated to the Early in the Migration Period, a large longhouse Iron Age shows that 44 per cent (38 out of 88 farms) was built at site 1. It was 40.5 metres long, 9.5 metres are located on farms with division names. The high- wide and covered an area of approximately 380 square est frequency is related to Pre-Roman Iron Age (55 per metres. It was exceptionally large, but followed the tra- cent). The material from the Merovingian Period shows ditional building style of the period, with separate in- 60 per cent, but is fairly modest as it only counts fi ve ternal sections for living space and holding animal hus- sites. The sites dated to the Migration Period that by bandry. Signifi cant resources must have gone into the far make up the largest body, appear in 38 per cent on construction. The longhouse was destroyed in a severe farms that hold such names, including minor sites. fi re around 600 AD. Shortly after, a male was buried in a mound located in the section where the living space Viking Age 2 of 9 = 22 per cent in the house had been. The grave goods indicate a war- Merovingian period 3 of 5 = 60 per cent Migration period 15 of 39 = 38 per cent rior with reasonable wealth and status (C 57449). The (included Vestre Ringdal) location of the inhumation seems to be deliberate and Roman Age 7 of 15 = 46 per cent symbolic. The body was laid with the head facing north Pre-Roman Age 11 of 20 = 55 per cent and furnished with a whole range of weapons, with Sum 38 of 88 = 43 per cent the exception of a lance (cf. Jørgensen 1999, 164–166; Grindkåsa 2012).The discovery of a brooch plate tells This demonstrates that a total of ten sites out of that he was buried with a cape, and most likely a ban- 17 larger settlements (59 per cent) that were aban- doleer was attached by a so-called skjoldtorn (shield- doned during or prior to the Migration Period, includ- thorn) brooch, which is a relatively rare fi nd in Nor- ing western Ringdal, are located at farms that carry wegian contexts. Only eight of such brooches are division names. This is the case for six settlements known from Norway, all of which have been found in from the Migration Period and four from the Pre-Ro- the eastern regions, and also from the Åker fi nd (Sol- man period. berg 2000, 192). This type of brooch was more com- This suggests that larger abandoned settlements are mon in and in Western Europe during the greatly over-represented at farms that hold division Migration and Merovingian Periods (Koch 1977; Reiss

Fig. 5. Site 1 Auli. Arialphoto, Tom Hei- breen, E18 project, MCH, .

8 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 6. Site 1 Auli. Location compared to the historical settlement structure.

1994; Grindkåsa 2012). The man was buried with three the names Auli and Aulerød may signal that the latter knives, a double-edged sword with damascus blade, was once a part of Auli. The name probably refers to probably an import, and a shield, amongst other arte- soil condition; Ölfvin, alfr m. gravel, vin, meadow (NG facts. The grave also contained the head of a horse and VI, 229). Therefore, the initial division of the farm ter- an amulet. The grave goods suggests that a person of ritory might have gone between Låne and Auli, and a high social standing was buried here and is an early the secondary between Auli and Aulerød. In 1390, the example of a high-status grave, inspired by Continen- area was split up into seven independent holdings be- tal models, which occurs more frequently during the longing to the monastery. Auli is recorded as the west- Merovingian Period. ern, northern and eastern Auli in the Red Book in the There are traces of other buildings at the site, indi- 1390s, and Låne as northern, southern and southern- cating a settlement. The site then seems to have been most Låne, while Aulerød was one holding at that time transformed to a grave fi eld comprising up to ten (RIB, 204f). As far as we know, the seven units had in- mounds. By mid-19th century, more than six grave dividual settlements, as is later recorded (Fig. 6; 7). mounds were still preserved in the landscape (Nicolay- Halldis Hobæk has examined the medieval property sen 1866, 138) (Fig. 6). structure in the area close to the site. The polygons on Looking at the topographical situation in the 6th to Fig. 8 represent concentrations of aristocratic, monas- 7th centuries, when the sea-level was higher, it is clear tic and royal land in the late Middle Ages. The largest that the farm territory of the abandoned site was con- one is the reconstructed royal land around Sem. It con- strained by two rivers and the sea. This coincides with sists partly of royal and monastic land. A large estate the historical farm’s boundaries. The site is located seems to have been separated from the royal manor exactly where the boundaries of three historical farms and subjected to the monastery of St Olav in Tønsberg met; namely Låne, Auli and Aulerød. All three farms during the mid-1100s. were later parts of the same large estate surround- A contemporary saga records that the church of St ing the manor of Sem (see below). The similarity in Mikael situated in Tønsberg in 1190 had landed prop-

RURALIA IX 9 Frode Iversen, Big bang, lordship or inheritance? ??–??

holdings. It is diffi cult to date the stages in this proc- ess. One of Norway’s largest estates from the Viking Period and the Middle Ages is situated here. The divi- sion of the farm territory into minor farms may well have been connected to the development of this estate in an early phase. I interpret this accordingly: the Auli farm was divided into to separate units (Låne and Auli) and the settlement was abandoned around 600 AD. A grave-fi eld, that has left traces of about 6–10 graves, was then established and continued in use for some generations. If the farm was split in two units of equal stature, then the grave-fi eld could be linked to two gen- erations of landholders who were buried at the site of their ancestral settlement. This practice could then have lasted for three to fi ve generations, indicated by the number of graves. This corresponds to around 60 to 100 years, with a hypothetically length of each gen- eration of approximately 20 years (Skre 1997). There are, however, some elements that go against such an interpretation. The name Låne, most likely stems from the word lán, probably meaning rented land (NG VI, 212). Oluf Rygh has proposed that the word lán is connected to the transfer of private land, in the sense of ‘rented land’, meaning land used in exchange Fig. 7. Site 1 Auli. Location compared to settlement in 1390 AD. 1) Nor- for a fee. If this name can be traced back to a primary dre Låne, 2) Søndre Låne, 3) Sønste Låne, 4) Aulerød , 5) Vestre Aulin, division, Låne must have been rented land at an ear- 6) Søndre Aulin, 7) Østre Aulin. ly stage, and thus belonged to a larger landowner. If this is the case, it may perhaps be more likely that the erty, and the income sustained the monastery of St grave-fi eld represents the same line of landholders. In Olav, indicating an earlier royal ownership. The Church this sense, the placing of the graves close to the farm of St Mikael was closely tied to the crown, being a royal boundary could have a double symbolic connotation. chapel. In the 1390s the monastery of St Olav owned Besides reminding the people using the land of Låne a total of 104 farms/or parts of farms in Vestfold (RB, of their previous landlords, the graves served to con- 204ff; Eriksson 1993, 103). The greatest concentration nect and keep the memory of the ancestral farmstead of these was in close proximity to the royal manor Sem. alive for some generations. The close ties between the monastic estate and the This is not the only example that may illuminate royal manor seem to have continued through the Mid- such processes in the landscape in Vestfold, although dle Age. An account from 1576 shows that peasants at it is diffi cult to elaborate such a good case as for Auli. the monastery had to pay duties in labour at the royal The second site, Vestre Ringdal, is located further south manor from ‘from old times’ (NRR II, 192) (Fig. 9). (Fig. 10) and was excavated by Mari Østmo in 2005. The The renowned Oseberg mound (834 AD) is located settlement consisted of 21 houses: traces of 13 long- at the outskirts of this concentration. Several rich houses and eight minor buildings and six graves from and monumental mounds in the area indicate this as Pre-Roman Iron Age and a grave from the Migration Pe- a power centre from at least the beginning of the 9th riod. Nearly 30 cooking pits and 19 hearths were also century. The manor Sem is also recorded in narrative found (Fig. 11). The fi nds (C 55048–C55077) consisted sources as a royal villa allegedly at the time of King Har- mostly of pottery fragments from both the houses and ald Hairfair in the late 9th century. The distance from the graves. At least six of the longhouses are from the site 1 to the medieval manor house at Sem (Jarlsberg) Migration Period, but it is diffi cult to determine wheth- was 3.5 km. This reconstructed cluster of royal land er they were used contemporaneously. Gjerpe and Øst- consisted of c. 50 farms. It stretches from site 1 in the mo suggests that the settlement consisted of perhaps south-west to the Oseberg mound to the north-east two or three simultaneous holdings during the Migra- and represents a huge concentration of power in this tion Period (Gjerpe – Østmo 2008, 134f). area at least from the Viking Age. The settlement was abandoned around 600 AD. It was The case seems to provide evidence for a process located 120 metres north of the present farm-boundary where one farm territory was split into several minor separating western and eastern Ringdal, as shown on

10 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 8. Concentrations of royal, aristocratic and ecclesiastical land near site 1 in the late Middle ages. After Halldis Hobæk 2008.

RURALIA IX 11 Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 9. Concentrations of royal and ecclesiastical land near site 1 (marked with a ring) in the late Middle ages. After Halldis Hobæk 2008.

12 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Fig. 10. The location of the the areas of case studies. Background map from Google Earth.

Fig. 12. Ringdal was mapped in a scale of 1:4000 in the divided. The farmstead at eastern Ringdal was situated period 1807–1820 AD (Wasberg 1970, 116), showing near a small hill, which fi ts well with the term Upper. four settlements at Ringdal – one at western Ringdal In 1647, the holdings had similar rent, which indicates and three at Eastern Ringdal. The northern and south- an equal division of the farm. The boundary between ern settlements at Eastern Ringdal are much younger the eastern and western units is documented on the settlements and not relevant here. The two presumed map from the 1800s and is the same as it is today. The farmsteads from the Middle Ages were located a dis- boundary of the named farm that faces west could be tance of 250 metres apart. The farm is mentioned as old. A grave fi eld with four stone foundations dated to upper and lower Ringdal (not eastern or western) in the the Early Iron Age is located just 20 metres from this 1390s (RB, 31) and was, most likely, already physically boundary. Its location in the outfi elds and around

Fig. 11. Site 2, vestre Ringdal. Arial photo, Tom Heibreen, E18 project, MCH, Univer- sity of Oslo.

RURALIA IX 13 Frode Iversen, Big bang, lordship or inheritance? ??–??

Discussion

Research has evolved from focusing on only the farm settlements in the fi rst half of the 1900s, to look- ing at the whole farm territories also as production units, especially from the 1980s onwards, to better understand the economic and social relations between the farm units. Until the 1990s direct evidence of prehistoric settlement in Eastern Norway was scarce (Østmo 1991). Farm names, graves and stray fi nds were the most important sources for early settlement histo- ry. It has been proposed that the strong link between historic farms and pre-historic graves indicated set- tlement continuity (Pedersen 1999, c46). In the 1990s the potential of top soil stripping for the understand- ing of settlement history was recognised. Since then, there has been a steady enlargement of source mate- rial, with a discovery of c. 50 new sites every decade. There is reason to believe that this development will continue. Still, certain tendencies can be observed in the available material already at this stage.

Fig. 12. Site 2, Vestre Ringdal, Location compared to the historical set- Working with this paper, I have examined the total tlement structure. Background map: Marie Ødegaard 2008. body of evidence with regard to early settlement in South-East Norway, trying to shed some new light on the three main hypotheses that have been in the fore 125 metres from the closest main road over the Ra, of discussion in Scandinavian research. The new evi- adds strength to the hypothesis that the grave-fi eld was dence demonstrates that the development in Eastern constructed to mark the boundary between the farms Norway is far more dynamic than has been presumed. Rødbøl and Ringdal (Ødegaard 2007, 77). The changes during the transition from Early to Late The abandoned settlement was located in the for- Iron Age are especially signifi cant. ested outfi eld area of the 1800s. The path road to the An assumption with regard to this work is that many farmstead at western Ringdal about 500 metres to the abandoned settlements are situated on farms having north passed the abandoned site. The distance to the division names, and that this material has a potential farmstead at eastern Ringdal was about 700 metres. It for illuminating processes concerning farm divisions. is not known how the arable land was organised in the Overall, the farm divisions may indicate intensifi cation Migration Period, but the site was located in the out- in fi eld structure and settlement. Possible explanations skirts of the infi eld as they are recorded on the map. If for this could be population increase and new modes this refl ects an earlier situation, it would suggest that of operation, with a shift in balance between animal the settlement had a good location in the transitional husbandry and arable cultivation. The quantitatively zone of infi eld to outfi eld. Studies of landownership proven macro conditions have been strengthened by in the region in the Middle Ages, indicate that Ringdal using two more qualitatively oriented case studies on was not a part of a larger medieval estate, like Auli a micro level. They both support a theory that there is (Ødegaard 2007, 34). a strong connection between abandoned settlements In my view this indicates that the settlement was and processes of farm divisions. My review shows that abandoned because of a split in a larger farm territory, also in East Norway, a trend of abandonment occurs and the establishment of two new separate settlements. far more frequently during the Migration Period than It is possible that a so-called odalskifte is relevant to in the other periods of the Iron Age. this case. The farm could easily have been split be- None of the larger abandoned settlements that have tween two brothers without any interference of mano- been excavated seems to have been abandoned during rial structures. It is therefore likely that the Migration the Roman Period or Viking Age. They were, however, Age farm may have consisted of two units at the time abandoned both during the Pre-Roman and Migration of abandonment. During the Middle Ages the farm Periods. Both of the case studies relate to Migration consisted of two units with split infi elds and separate period sites. There are some source critical challenges settlements. The abandonment may have been a result tied to the dating of the abandonment, but the context of a division of farmland territory around 600AD. of the sites is good. The data implie that the abandon-

14 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

ment occurred around 600 AD, and not in the decades with reduced activity during the Migration Period (Goll- following 536 AD. It is uncertain whether these cases wizer 2012; Storrusten – Østmo 2012). This complex are representative, but they do not support Gräslund’s development makes it diffi cult to explain the changes hypothesis. Other hypotheses should therefore be con- using a ‘one-factor’ theory based on a single natural sidered. disaster, such as the one in 536 AD (Gjerpe 2012). Site I is located close to what is considered as one The three hypotheses should perhaps be seen ho- of Norway’s largest estates during the Viking Period listically. If the minor farms were abandoned because and Middle Ages. This case study is interesting as it of a climatic crisis, it could have caused new grazing shows that a large building burned down and that a lands in dense settled areas like the Ra. An easier ac- settlement cluster was abandoned. The construction cess to manure could have made a more intensive use of a grave mound at the site immediately followed this of arable farming possible, which again resulted in event, in which a weapon-equipped person of high so- more farms being divided over time. The development cial standing was buried. Over time, this area devel- is obviously complex and multiple factors need to be oped into a grave-fi eld. The abandoned settlement was assessed to be able to draw any conclusions of cer- situated close to the boundary that came to separate tainty. The examined material contains elements that three historical farms. In turn, these formed a natu- support the crisis hypothesis. It seems likely that a re- ral farm territory during the 6th and 7th centuries. In organisation of the settlement accelerated at the end 1390, the area was split into seven smaller units that of the Migration period or in the early decades of the were all part of a larger estate holding. This could be Merovingian Period. On a practical level, this resulted interpreted as a sign of intensifi cation and reorganisa- in, amongst others, a division of farmland. The mate- tion of the settlement under the auspices of a central rial gives few indications of the driving forces that led landowner. On the other hand, Case 2 shows an equal to such development. It could be about demographic division of a farm territory. The abandoned settlement relations, where the potential for grain cultivation was consisted of traces of 21 buildings and at least two more exploited in areas where this was possible. This simultaneous units during the Migration Period. The could have resulted in a growing surplus and, as such, settlement was therefore most likely moved when the created a foundation for emerging elites and minor farm was divided. kingships. Changes of hereditary rights may then have Could this development have anything to do with a been a strategy that was motivated by and stimulated change of hereditary rights? It can be observed that to this desired development of both larger estates and larger settlements were not abandoned during the Ro- split up settlements and farmland. man Period. On the other hand, this was the case in Pre-Roman and Migration Periods. The time lapse be- Summary tween these two periods is extensive and the interpre- tation is therefore rather uncertain. Still, it is not un- This article discusses the archaeologically excavated likely that the division processes during the Migration settlement sites in South-Eastern Norway in relation to Period may have been connected to hereditary norms. three hypotheses under discussion in Scandinavian ar- Prehistoric farming in Eastern Norway has also been chaeology. It is suggested that also processes of inher- archaeologically examined in recent years. The E18 itance and landownership is important to understand project in Vestfold has demonstrated the usefulness of the changing settlement pattern on the threshold of the combining scientifi c methods with traditional excava- Merovingian period. Changes of hereditary rights may tion of fossil traces of cultivation. Micro-morphologi- then have been a strategy that was motivated by and cal analyses have shown changes in fertilisation and stimulated to this desired development of both larger land exploitation over time. Botanical analyses of wood estates and split up settlements and farm land. show, among others, that forests were used as a re- source for gathering fodder. Altogether, the settlements Acknowledgement by the Ra were more signifi cant than has previously been understood (Mjærum – Gjerpe 2012a; 2012b). Sev- The article is partly based on a list over settlement eral sites indicate, however, decline already during the sites in Eastern Norway, which will be published in Migration Period. In Hørdalsåsen there are signs that full in The Museum of Cultural History’s program on cultivation terminated towards the end of the Roman Settlement archaeology (Martens in prep). The list is period, while pastures expanded during the Migration compiled by Jes Martens and updated in collaboration Period (Mjærum 2012). Traces of cultivation close to with Lars Erik Gjerpe and Margrete Simonsen. I am the settlement areas at Hesby and Østre Borge, seem thankful for them letting me use this in my article. to indicate stable settlements from the Bronze age to Thanks to Lars Erik Gjerpe and the E18 project for present times, but also shows a complex development, thought provoking and inspiring discussions on set-

RURALIA IX 15 Frode Iversen, Big bang, lordship or inheritance? ??–??

tlement structure these last two years, and Ingvild Øye Carlie, L. 1999: for reading the manuscript critically I would also like Bebyggelsens mångfald. En studie av södra Hallands järnå- to thank Jessica McGraw for input and translating this ldersgårdar baserad på arkeologiska och historiska källor. article. Acta Archaeologica Lundensia Series. Lund. Diinhoff, S. 2011: Chiefl y Manors and the Establishment of a Socially Hierar- Zusammenfassung chical Settlement Pattern in Western Norway. Arkæologi I Slesvig. Det 61. Internationale Sachenssymposion 2010. Ha- Dieser Artikel beschreibt die archäologisch ausge- derslev, Danmark, 211–222. Neumünster. graben Siedlungen in Süd-Ost-Norwegen in Bezug auf Edblom, L. 2004: drei Hypothesen zur Diskussion in die skandinavische Långhus i Gene: teori och praktik i rekonstruktion. Studia Archäologie. Es wird vermutet, dass auch Prozesse der archaeologica Universitatis Umensis 18. Umeå. Vererbung und Zugehörigkeit zu Land wichtig ist, die Ericsson, A. 2001: wechselnde Siedlungsstruktur an der Schwelle der Me- Möre mellam järnålder och medeltid – omvandlingen av ett rowingerzeit zu verstehen. Änderungen des Erbrechts agrarlandskap. Möre – historien om ett Småland, 367–414. kann dann eine Strategie gewesen, die durch motiviert Kalmar. war und stimuliert auf diese gewünschte Entwicklung Erikson, O. – Strid, J. P. 1991: der beiden größeren Gütern und aufteilen Siedlungen Runstenar. Malmö. und Ackerland haben. Eriksson, J. E. G. 1993: Tønsberg og omlandet – Olavsklosterets økonomiske grunn- Résumé lag i Vestfold. Seminaret ”Kloster og by” 11.–13. november 1992: omkring Olavsklosteret, premonstratenserordenen Cet article discute les sites archéologiques fouillés og kloster vesenet i middelalderen (eds J. E.G. Eriksson – dans le règlement du Sud-Est de la Norvège par rapport K. Schei), 94–111. Tønsberg. à trois hypothèses en cours de discussion en archéolo- Fabech, C. – Ringtved J. 1999: gie scandinave. Il est suggéré que aussi des processus Settlement and Landscape. Proceedings of a conference in de succession et de propriété à la terre est important Århus, Denmark, May 4–7 1998. Århus. de comprendre la structure de l'habitat évolue sur Frölund, P. 2007: le seuil de l'époque mérovingienne. Changements Gamla Uppsala under äldre jernålder. Land och samhälle i de droits héréditaires peuvent ensuite avoir été une förändring. Uppländaska bygder i ett långtidsperspektiv. Ar- stratégie qui a été motivée par et stimulé à cette évolu- keologi E4 Uppland-studier 4, 361–378. Uppsala. tion souhaitée des deux domaines de plus en séparer G = Den eldre Gulatingslova. Norrøne tekster 6, eds B. Eit- les colonies et les terres agricoles. hun, M. Rindal og T. Ulset, 1994. Oslo. Gansum, T. 1995: Jernaldergravskikk i Slagendalen – Oseberghaugen og stor- haugene i Vestfold – lokale eller regionale symboler. Upublis- hed disertation in archaeology. University of Oslo. Bibliography Gjerdåker, B. 2001: Til odel og eige – odels- og åsetesrettane gjennom eit mille- Ahlkvist, H. B. 2002: nium, med vekt på dei siste 250 åra. NILF-rapport / Norsk Hällristarnas hem – gårdsbebyggelse och struktur i Pryssgår- institutt for landbruksøkonomisk forskning. Oslo. den under bronsålder. Riksantikvarieämbetet. Arkeologiska Gjerpe, L. E. 2008: undersökningar 42. Stockholm. E18-prosjektet Vestfold, Bind 3, Hus, boplass- og dyrknings- Bratt, P. 2010: spor. Varia, vol. 73. Oslo. Makt uttryckt i jord och sten: stora högar och maktstrukturer Gjerpe, L. E. 2012: i Mälardalen under järnåldern. Stockholm studies in archae- Fra natur til landbrukslandskap. Oppsummering Gulli–Lan- ology bd. 46. Stockholm. gåker. E18-prosjektet Vestfold bind 3. Oslo. Brink, S. 2002: Gjerpe, L. E. – Østmo, M. 2008: Law and legal customs in Viking age Scandinavia. The Scan- Kapittel 3. Ringdal 13 – Hus fra romertid-merovingertid og dinavians from the Vendel period to the tenth century – an graver fra førromersk jernalder. E18-prosjektet Vestfold, ethnographic perspective (ed. J. Jesch), 87–127. Woodbridge. Bind 3, Hus, boplass- og dyrkningsspor. Varia, vol. 73, 39– Büngten et al. 2011: 142. Oslo. 2500 Years of Climate Variability and Human Susceptibility, Gollwizer, M. 2012: Science 331, no. 6017, 578–582. Graver, brønner og bosetningsspor fra bronsealder til mid- Bårdseth, G. A. 2007: delalder på Hesby (lokalitet 13), in: A. Mjærum – L.E. Gjerpe E6-prosjektet Østfold, Hus, gard og graver langs E6 i Sarps- (eds): Jordbruksbosetning og graver i Tønsberg og . borg kommune, vol 1–3. Varia vol 65–67. Oslo. E18-prosjektet Vestfold bd.1. Oslo.

16 RURALIA IX Frode Iversen, Big bang, lordship or inheritance? ??–??

Grindkåsa, L. 2012: Løken, T. – Pilø, L. – Hemdorff, O. 1996: Boplasspor og grav fra romertid–merovingertid på Jarlsberg Maskinell fl ateavdekking og utgravning av forhistoriske jord- og Tem (lokalitet 8, 9 og 10), in: A. Mjærum – L. E. Gjerpe bruksboplasser: en metodisk innføring. AmS-varia vol 26 (ed. (eds): Jordbruksbosetning og graver i Tønsberg og Stokke. K. Griffi n). Stavanger. E18-prosjektet Vestfold bd 1. Oslo. Løken, T. 2001: Gräslund, B. 2007: Oppkomsten av den germanske hallen – hall og sal i eldre Fimbulvintern, Ragnarrök och klimatkrisen år 537–537 e. jernalder i Rogaland, Viking, 49–86. Oslo. Kr. Saga och Sed, Kungl. Gustav Adolfs Akademiens årsbok Martens, J. in prep: 2007, 93–123. Uppsala. Faglig program for bosetningshistorie, Museum of Cultural Göthberg, H. 2000: history, University of Oslo. Bebyggelse i Förändring. Uppland från slutet av yngre bron- Martens, J. 2007: sålder till tidig medeltid. Occasional Papers in Archaeology Kjølberg søndre – en gård med kontinuitet tilbake til eldre 25. Uppsala. jernalder? Varia 2007, vol 62. 89–109. Oslo. Göthberg, H. 2007: Mjærum, A. 2012: Mer än bara hus og gårdar. Hus och bebyggelse i Uppland. Dyrkningsspor og fegate fra eldre jernalder på Hørdalen (lo- Delar av förhistoriska sammanhang. Arkeologi E4 Uppland, kalitet 51), in: L. E. Gjerpe – A. Mjærum (eds.): Dyrking, bo- vol. 3, 403–447 (ed. H. Göthberg). Societas Archeologica Upp- setninger og graver og i Stokke og . E18-prosjektet saliensis. Uppsala. Vestfold bd 1. Oslo. Helgesson, B. 2002: Mjærum A. – Gjerpe, L. E. (eds) 2012a: Järnålderns Skåne: samhälle, centra och regioner. Uppåkra- Jordbruksbosetning og graver i Tønsberg og Stokke. E18-pro- studier 5, Acta archaeologica Lundensia 38. Lund. sjektet Vestfold bind 1. Oslo Historia de profecitione Danorum in Hierosolyman = Jorsals- Mjærum A. – Gjerpe, L. E. (eds.) 2012b: ferda åt danene, translated by B. Svare, 1934. Norrøne Bo- Dyrking, bosetninger og graver og i Stokke og Sandefjord. kverk nr. 31. Oslo. E18-prosjektet Vestfold bind 2. Varia. Oslo. Hobæk, H. 2008: Myhre, B. 1972: Kongsgården Sem i Vestfold: en arkeologisk analyse av hoved- Funn, fornminner og ødegårder. Jernalderen bosetning i Høy- gård og gods i middelalder og yngre jernalder. Unpublished land Fjellbygd. Stavanger museum skrifter 7. Stavanger. master thesis, University of Bergen. Myhre, B. 1992: Holst, M. 2004: Borre – et merovingertidssenter i Øst-Norge. Økonomiske og The Syntax of the Iron Age Village: Transformations in an or- politiske sentra i Norden ca 400–1000 e. Kr. – Åkerseminaret, derly community. Århus University. Hamar 1990 (eds. E. Mikkelsen – J. H. Larsen). Universitetets Iversen, F. 2008: oldsaksamlings skrifter. Ny rekke 13, 155–179. Oslo. Eiendom, makt og statsdannelse : kongsgårder og gods i Hor- Myhre, B. 2002: daland i yngre jernalder og middelalder. UBAS, Nordisk 6. Landbruk, landskap og samfunn, in: B. Myhre – I. Øye: Nor- Bergen. ges Landbrukshistorie bd. 1, Jorda blir levevei, 4000 f.Kr.– Iversen, F. 2009: 1350 e.Kr., 12–213, Oslo. Den nyere norske forskningen om jordegods i vikingtid og NG = Norske Gaardnavne, I–XVIII, ed. O. Rygh et al. Kra. tidlig middelalder. En sammenligning med undersøkelsene 1897– av manor- og Grundherrschaft-systemer i Vest-Europa. Den tapte middelalder? Middelaldernes sentrale landbebyggelse Nicolaysen, N. 1866: (eds. Martens J. – Martens, V. – Stene, K). Varia 71, 59–70. Norske fornlevninger. En oplysende fortegnelse over Norges Oslo. fortidslevninger ældre end reformationen og henførte til hver sit sted. Kristiania. Iversen, T. 1997: Trelldommen norsk slaveri i middelalderen. Skrifter, Histo- NRR = Norske Rigs-registranter (utg. av C.C.A. Lange et al. risk institutt, Universitetet i Bergen 1. Bergen. vol. IV). Christiania 1861–1891. Jørgensen, A. N. 1999: Olsen, M. 1926: Waffen und Graber: typologische und chronologische Studien Ættegård og helligdom : Norske stedsnavn sosialt og religi- zu skandinavischen Waffengrabern 520/30 bis 900 n. Chr. onshistorisk belyst. Instituttet for sammenlignende kultur- Nordiske Fortidsminder ser. B Vol. 17. Copenhagen. forskning Serie A: Forelesninger 9a. Oslo. Kleven, H. I. 2001: Pedersen, E. A. 1990: Norrøn arverett og samfunnsstruktur. Oslo. Rydningsrøysfelt og gravminner – spor av eldre bosetnings- Koch, U. 1977: strukturer på Østlandet. Viking, tidsskrift for norrøn arkeo- Das Reihengräberfeld bei Schretzheim. Germanische Denk- logi, bd LIII. Norsk Arkeologisk Selskap, 50–66. Oslo. mäler der Völkerwanderungszeit. Serie A / Römisch-germa- Pedersen, E. A. 1999: nische Kommission des Deutschen Archäologischen Insti- Transformations to sedentary farming in eastern Norway: tuts, vol. 13. Berlin : Mann. AD 100 or 1000 BC? Settlement and Landscape Proceedings

RURALIA IX 17 Frode Iversen, Big bang, lordship or inheritance? ??–??

of a conference in Århus, Denmark, May 4–7 1998, C. Fabech http://arkeologi.blogspot.com/2005/03/tunformer-og-bebyg- – J. Ringtved (eds.), Jutland Archaeological Society. Gylling. gelse-i.html Petersson, M. 2006: Söderberg, B. 2005: Djurhållning och betesdrift : djur, människor och landskap Aristokratiskt rum och gränsöverskridande. Järrestad och i västra Östergötland under yngre bronsålder och äldre jär- Sydöstra Skåne mellan region och rike 600–1100. RAÄ Ar- nålder. Stockholm – Uppsala. keologiska undersökningar. Skrifter 62. Stockholm. Petersson, M. 2011: Tacitus, Cornelius 98. Agricola ; Germania : lateinisch und The Early Iron Age Landscape. Social Structure and the Or- deutsch / Cornelius Tacitus , translated Alfons Städele 2001. ganisation of Labour. Arkæologi I Slesvig. Det 61. Internatio- Zürich. nale Sachenssymposion 2010. Haderslev, Danmark, 249– Verhulst, A. 1995: 268. Neumünster. Aspekte der Grundherrschaftsentwicklung des Hochmittelal- RB = Røde bok. Biskop Eysteins Jordebog (Den røde Bog) : ters aus westeuropäischer Perspektive. Grundherrschaft und Fortegnelse over det geistlige Gods i Oslo Bispedømme om- bäuerliche Gesellschaft im Hochmittelalter (ed. W. Rösener). kring Aar 1400. Det norske historiske Kildeskriftfond 10 (ut- Wasberg, G. C. 1970: gitt ved H.J. Huitfeldt 1879). Christiania. Brunlanes. En bygdebok. Bind 1. Bygdehistorie. . Reiss, R. 1994: Zachrisson, T. 1994: Der merowingerzeitliche Reihengräberfriedhof von Westheim The Odal and its manifestations in the Landscape, Current (Kreis Weissenburg-Gunzenhausen) : Forschungen zur früh- Swedish Archaeology 2, 219–238. Stockholm. mittelalterlichen Landesgeschichte im südwestlichen Mittel- franken. Wissenschaftliche Beibände zum Anzeiger des Ger- Zachrisson, T. 2011: manischen Nationalmuseums, Bd. 10. Nürnberg. Property and Honour – Social Change in Central Sweden, 200–700 AD Mirrored in the Area around Old Uppsala. Ar- Reitan, G. 2009: kæologi I Slesvig. Det 61. Internationale Sachenssymposion Rapport fra arkeologisk utgraving. Moi (Gnr. 12/1, 2, 3 og 4), 2010. Haderslev, Denmark, 141–156. Neumünster. Bygland kommune, Aust-Agder. Kulturhistorisk museum. Universitetet i Oslo. Zimmermann, W. H. 1998: Pfosten, Ständer und Schwelle und der Übergang vom Pfo- Reynolds, A. 1999: sten- zum Ständerbau – Eine Studie zu Innovation und Be- Later Anglo-Saxon England: life and landscape. Stroud. harrung im Hausbau. Zu Konstruktion und Haltbarkeit prä- Robberstad, K. 1967: historischer bis neuzeitlicher Holzbauten von den Nord- und Odelsrett. KLNM XII, 493–499. Ostseeländern bis zu den Alpen. Probleme der Küstenfor- Rösener, W. 1995: schung im südlichen Nordseegebiet 25, 9–241. Einführung. Grundherrschaft und bäuerliche Gesellschaft Ødegaard, M. 2008: im Hochmittelalter. Veröffentlichungen des Max-Planck-Insti- Graver og grenser: territoriell organisering av gårdene i jern- tuts für Geschichte 115, 9–15. Göttingen. alderen i Søndre Vestfold. Unpublished master thesis, Univer- Skre, D. 1997: sity of Bergen. Haug og grav. Hva betyr gravhaugene, in: Middelalderens Ødegaard, M. 2010: symboler (eds A. Christensson – E. Mundal – I. Øye). Kultur- Graver og grenser. Territoriell inndeling av jernalderens jord- tekster 11, 37–52. Bergen. brukslandskap i Vestfold. Primitive tider 12 2010, 27–39. Skre, D. 1998: Oslo. Herredømmet bosetning og besittelse på Romerike 200–1350 Østmo, E.1991: e. Kr. Acta Humaniora 32. Oslo. Gård og boplass i østnorsk oldtid og middelalder Aktuelle Streiffert, J. 2005: oppgaver for forskning og forvaltning, Varia 22, Universite- Gårdsstrukturer i Halland under bronsålder och äldre jär- tets oldsakssamlings skrifter. Oslo. nålder. GOTARC, Series B, Gothenburg archaeological theses Øye, I. 2002a: 66. Mölndal, Göteborg. Vestlandsgården – fi re arkeologiske undersøkelser, L. Juls- Sorrusten, E. – Østmo, M. 2012: hamn, R. Bade, K. A. Valvik, J. Åstveit (ed. I. Øye). Arkeolo- Depot, bosetningsspor og dyrkningsspor fra neolitikum giske avhandlinger og rapporter fra Universitetet i Bergen 8. til middelalder på Borge østre (lokalitet 22), in: A, Mjærum Bergen. – L. E. Gjerpe (eds): Jordbruksbosetning og graver i Tønsberg Øye, I. 2002b: og Stokke. E18-prosjektet Vestfold bd 1. Kulturhistorisk Mu- Landbruk under press, 800–1350. In: B. Myhre & I. Øye: Jor- seum, Fornminneseksjonen. Oslo. da blir levevei – 4000 f. Kr. –1350 e.Kr. Norges landbrukshis- Stylegar, F. A. 2005: torie 1, 214–453. Oslo. Tunformer og bebyggelse i middelalderen. From his blog:

Dr. Frode Iversen, Museum of Cultural History, University of Oslo. [email protected]

18 RURALIA IX