'The time has come,' the Walrus said,
'To talk of many things:
Of shoes — and ships — and sealing-wax —
Of cabbages — and kings —
And why the sea is boiling hot —
And whether pigs have wings.'
Lewis Carroll: The Walrus and the Carpenter

22 mars 2007

Late-Medieval Aristocratic Landownership in Shetland


’The lords of Norroway’ are mentioned in several Shetland sources from the sixteenth and seventeenth centuries. These ‘lords’, aristocratic families based in (Western) Norway, were inheritors of the vast estate that a lady of Shetland origins, Herdis Thorvaldsdatter, had amassed until her death in 1363. The present paper takes a closer look at this estate.

Herdis was the daughter of herra Thorvald Thoresson, the king of Norway’s syslumaðr in Shetland in the 1290s, in his second marriage to Ragndid. Thorvald and Ragndid had two children, a son, Thorvald, and Herdis. The latter was probably born around 1305. Herdis first enters the sources in 1328, when she met with other prominent men and women to assess a dowry (DN II, 165). Thus, she must have reached the full legal age of twenty at that time.

Herdis made one, probably two strategic marriages. Her first, documented marriage was to Svein Sigurdssson, who is mentioned as regional treasurer in Bergen 1328-1329. Svein died in 1332. Herdis’ second marriage, as suggested by the historian P. A. Munch in 1862, was probably to the Norwegian nobleman, Bjarne Erlingsson, who married an unnamed, prominent woman in 1343. This earned Herdis the title of ‘fru’ (lady), which is used in the later sources (after Øye 2002:88). Bjarne was the only son of the drottsete Erling Vidkunsson, that is the highest royal official of the time, who also owned the Bjarkøy estate in Northern Norway. Bjarne Erlingsson, who was knighted on his own account, died in 1349, leaving Herdis a widow for the second time.

Herdis’ estate
Herdis died in 1363. According to an Icelandic annal, it happened in Copenhagen, where lady Herdis was travelling with the Norwegian royal court (Isl. ann., p. 360, which also, without any support in other sources, states that there was foul play involved). She was without surviving siblings, children, or siblings’ children (DN II, 375, 1363; also Sollied 1942), a destiny most likely met by many in the years immediately following the Black Death. Taking Norwegian rules of inheritance into account, this means that she must have inherited both from her father and from her brother. Furthermore, she must have received substantial gifts from her two husbands, both of whom were wealthy landowners.

In the case of Svein Sigurdsson, Herdis’ dowry and dower (i.e. the values brought by Herdis into the marriage and the values pledged by Svein in response, respectively) amounted to more than 300 forngilde, Norwegian marks. In addition, Herdis got substantial gifts from Svein (Øye 2002:87-88). Among the gifts were also properties in Shetland (DN I, 220, 1332).

We don’t know about the gifts presumably given by Bjarne Erlingsson to Herdis, but they were probably substantial.

In her later years, Herdis also aquired a number of properties in Shetland, thus adding to her fortune. Two property deeds are preserved, one dealing with properties in Walls (Shetland documents 1195-1579, No 9; cf DN III, 284, 1355), and one with properties in Unst (Shetland documents 1195-1579, No 11; cf DN III, 310, 1360). Her Shetland estate seems to have amounted to more than 1300 Shetland marks (see below).

Being without direct heirs, Herdis decided on her deathbed that both her inherited estates and the properties she had bought were to be used for the foundation of a Cistercian monastery. In case the monastery was not built, her estate would pass to the king (Øye 2002:89).

Why this monastery did not come into being, we do not know. Already in her year of death, 1363, the lawman in Bergen enquires into the relationship between the deceased and herra Sigurd Hafthorsson, and it was concluded that the father of the latter, the mighty Hafthor Jonsson of Sudreim (Sørum) in Eastern Norway, who had been married to Agnes, the daughter of king Håkon the fifth, was the brother of Herdis’ mother, lady Ragndid. Thus, Sigurd Hafthorsson and his brother, Jon, being the closest in kin, were granted Herdis’ estate (DN II, 375, 1363; cf Øye op cit). Her properties thus passed to her mother’s side of the family.

The king’s claim to Herdis’ estate is never mentioned again in the preserved sources. Soon after, Earl Henry Sinclair seems to have set out deliberately to incorporate Shetland within his Orkney earldom, and in 1379 had to promise king Håkon the sixth that he would not alienate or sell any properties of the earldom away from the king (Crawford & Ballin Smith 1999:18-19). Some years later, we learn that Malise Sperra, a claimant of the earldom family and most likely also a royal ‘foud’, had seized the properties owned by Herdis Thorvaldsdatter, but that these were now restored to Sigurd and Jon Hafthorsson (Crawford 1983:38, Shetland documents 1195-1579, No 13; cf DN I, 501, 1386). Presumably, the king dropped his claim to Herdis’ estate (Øye 2002:89).

When Sigurd Hafthorson died, his estate passed to his daughter. When she died, it passed to her son, Sigurd Jonsson of Sørum and Giske (the latter farm situated in Møre, Western Norway), then the drottsete and the greatest landowner in Norway around the middle of the fifteenth century.

We do not have many sources from which to learn of the handling of the Shetland part of the vast Sørum-Giske estate in the fifteenth century. However, in 1485 we hear of plundering in Shetland by ’the lords of Norway and their agents’ (quoted after Crawford & Ballin Smith 1999:21). This may reflect a reaction to the arrival in the north of Lord Henry Sinclair in 1480, as suggested by Barbara Crawford (loc sit). Under any circumstances, it is a clear sign that the absentee landlords were (still) prepared to defend their properties by any means necessary.

It was not until well after the pledge of Shetland to Scotland (1469), in 1490, that Sigurd Jonsson’s estate was formally divided. By that time, 25 years had passed since the death of the last owner of the undivided estate, Hans Sigurdsson, the son of Sigurd.

The tripartite division of the estate
There were three parties to the division of Sigurd’s estate in 1490. The inheritors were Otte Matsson Rømer, Alv Knutsson (tre roser), and Arald Kane, the latter acting on behalf of two sisters, Sigrid Erlendsdatter and Sigrid Erlendsdatter (Shetland documents 1195-1579, No 30; cf DN VIII, 426).

The deed of division specifying which parts of Sigurd Jonsson’s estate befell each of the three benefactors, is the first listing of the properties that lady Herdis Thorvaldsdatter had amassed in Shetland. Although lacking in details, the list gives us insight into the basic structure of her Shetland estate.

Alv Knutsson seems to have inherited the most important Norwegian properties, Giske and Sudrheim. In Shetland he got Vaila and ‘all the lands’ in Walls (Vogafiordwngh), Foula, properties in Sandness and Northmavine, as well as ‘all the lands’ in Unst and Fetlar. Otte Matsson’s part consisted of the Kvåle estate in Sogn, and in Shetland, Papa (Stour), Wethersta and Busta in Delting, ‘all the lands’ in Northmavine (Mawedes ‘otting’), and ‘the lands’ in Yell. The Dønnes estate in Northern Norway featured prominently in the part allotted to Arald Kane. He also got properties in Shetland: Noss ‘with all the properties laid under it’, ‘the lands’ in Bressay, Ustaness in Whiteness, and properties in Tingwall, Delting, Lunnasting, Nesting and Whalsay.

The standard phrase used in Scottish records when referring to these three estates is ‘Vaila and Vailaguids’, ‘Papa and Papaguids’, and ‘Noss and Nossguids’, ‘guids’ being the Norwegian term gods, i.e. estate.

Beyond the basic composition of the estate, the 1490 document leaves much to be desired. To get any more details, we have to consult later sources dealing with each of the three parts. The Vaila part is best covered. Gørvel Fadersdatter (Sparre of Hjulsta and Angsö), Alv Knutsson’s great granddaughter, inherited the estate. Fru Gørvel’s huge estate consisted of the majority of the old Sudrheim estate, the Bjarkøy estate, and the Giske and Finne estates (Bjørkvik 1958:64). In 1582 she made over her land in Shetland to the king of Denmark-Norway, and a list of her properties was drawn up (Shetland documents 1580-1611, No 42; Daae 1895). This gives detailed information about the properties in Shetland, i.e. their location, value and tenancies.

As far as the sources go, the Vaila estate was left unchanged between 1490 and 1582. We know, however, that both Knut Alvsson and Karine Alvsdatter, both of whom controlled Alv Knutsson’s estate before it passed to lady Gørvel, did sell off lands in Norway. Thus, Knut Alvsson sold the Bjørø estate in Trøndelag (DN I, 995 1498), while Karine Alvsdatter sold the Æri estate (DN IV, 1114, 1539). Furthermore, from a document dated 1531 (?) we learn that the Shetland part of the deceased Alv Knutsson’s estate at that time was owned by Ture Jonsson (tre roser), who also administered the estate in Western and Northern Norway on behalf of lady Gørvel (DN XIV, 540). But none of the sources passed down to us tell of any transactions involving lands in the Shetland part of the estate.

There is a potentially more serious source-critical problem with the Papa part. In this case, the primary source is a charter of 1624 in favour of Gilbert Mowat which lists the lands in Northmavine (NAS, GD/248/395/2). The problem with this document is that one cannot be certain that all these lands were part of the original Papa estate, since the Mowat family might have aquired other lands by 1624. On the other hand, some properties must have been removed from the estate between 1490 and 1624. Busta and Wethersta in Delting belonged to the estate inherited by Otte Matsson in 1490, but were on other hands even in the late sixteenth century.

There is also a bond of surety of 1631 in favour of Christian Stewart, lessee of Papa and Papaguids (NAS, CS290/62, reg. 20). It does not list the lands, but it names various tenants on the lands. It is most likely that their addresses are those of farms in the Papa estate. There is a certain degree of congruence between the two lists. Unfortunately, there are no Yell lands in the 1624 list, and, even if some farms on this island are mentioned in the document from 1631, it seems impossible to reconstruct the Yell part of the estate. All the lands in Papa Stour itself belonged to the estate, as witnessed by several sources (Crawford & Ballin Smith 1999; Smith 2000).

For the Noss estate, there is a feu charter in favour of Thomas Leslie of Burwick, dated 1664, listing ‘his udal landis eftermentioned callit and disingued (sic) lordis of Norrowayis land of Noss and Nossgoodis’ (NAS, RS44/4 folios 153r.-154v). This seems to be an exhaustive list of the lands belonging to the estate, although, again, we cannot be sure whether the Noss part of Sigurd Jonsson’s estate in 1490 still remained intact 175 years later, or if some lands had been sold off or, less likely in our opinion, new ones had been added.

As a general point, and given the historical context, it seems more likely that lands have been ceded from the original estate during the sixteenth and early seventeenth centuries, than the opposite. This point of view is perhaps strengthened by the way the estates of the ‘lords of Norroway’ finally got into Scottish hands. As for Papa, the rights of the absent landowners
were simply ignored by Gilbert Mowat and his son James who, as stated in a document written in the 1630s, ‘have intruded themselve in the lands called Papastoure and sindrie uther lands pertaining to the Lords of Norroway and keipis themselves in possesion thereof be bangsterie and oppression’ (quoted after Crawford & Ballin Smith 1999:36). The story at least for the Vaila estate was similar (Smith 1990:34). It should be added that, with the exception of Busta and Wethersta in Delting, none of the younger sources discussed above list properties that are incongruent with the, admittedly rather general, descriptions used in the deed of 1490. All in all, and excluding the lands in Yell, the available sources seem to give a rather comprehensive view of the estate amassed by lady Herdis.

Lady Herdis’ Shetland estate
With the above-mentioned source-critical reservations, we can reconstruct the estate owned by Herdis Thorvaldsdatter at the time of her death in 1363 (table 1).

In total, it comprises more than 1300 Shetland marks. The Vaila part is 470 marks, while Papa is 480, and Noss 370. Even if the Vaila and Papa parts seem equal in value, we must remember that no Yell farms are included in the list. Thus, the Papa part of the estate might have been considerably larger. Busta and Wethersta, mentioned as belonging to Otte Matsson’s part in 1490, are not included, either. The way they are mentioned in the deed of 1490, might seem to suggest that we are dealing with whole townships (for this interpretation, see Crawford & Ballin Smith 1999:31). However, other lands specified in the deed are clearly not whole townships. One can compare with Ustaness in Whiteness (provided that Ustaness is what is meant with ’Odwristanes’ in the older source), which belonged to the Noss part and in 1664 turns out to be a property of 21 marks out of a township valued at 72 marks. The 1490 document does not list any other properties in Whiteness and Weisdale, though, while the feu charter of 1664 lists seven additional, minor properties, which together make up for another 20 marks. The naming of Wethersta and Busta in 1490 may hint to similar, smaller clusters of lands centred on these two, probably larger, properties.

Accordingly, the Papa part might have been considerably larger, and thus significantly bigger than the Vaila part. However, Alv Knutsson, who inherited the Vaila part, seems to have got the better deal regarding the Norwegian part of the estate in 1490 (Crawford & Ballin Smith 1999:31), and the distribution of the Shetland lands might have been a way of evening out the differences. The reason for the comparatively minor size of the Noss part of the estate may seem more difficult to explain. The better interpretation, to us, is that the three parts in 1490 constituted three distinct and separate – and pre-existing – estates, formally and functionally similar to the Norwegian ones (below).

The undivided estate counted c. 180 properties in Shetland, not counting a couple of small islets and three seal waiths, and bearing in mind that we know next to nothing about the Yell part of the estate. 70 of these are valued at 6 marks or more. 6 marks seems a reasonable size for an average tenancy, cf the numbers in Gifford’s rental of 1716 (NAS, RH.9/15/176). As many as 45 of the properties in the estate constitute 50 per cent or more of the farms/townships where they are located.

Some notes by the superintendent in Bergen, Geble Pedersen, dating to c. 1550, state that ’18 marks burnt in land is what we call a ‘manswerch’ in Norway, but in Hieltland it is bigger’ (Shetland documents 1195-1579, No 91). A (post-medieval) ‘man’s work’ in Western Norway was counted as 4 laupsleie, and in the mid-seventeenth century 3 laupsleie was counted as a ‘full’ farm. Applied to Shetland, and keeping in mind that this is by no means an accurate comparison, this means that a farm valued at 13,5 marks equals a ‘full’ farm in Norway. Thus, lady Herdis’ Shetland estate equalled c. 100 ‘full’ Norwegian farms. But only 21 of her properties were actually valued at 13,5 marks or more.

In Shetland respects, this property of some 1300 marks was an estate of unrivalled size. For the sake of comparison can be mentioned that ‘the lands and barony of Brugh’ (Hew Sinclair) in 1607-1608 was valued at c. 330 marks (Shetland documents 1580-1611, Nos 436, 467), while a disposition in 1606 by Lawrence Sinclair of Gott to his son James Sinclair and James’ wife, Christiane, in fulfilment of a matrimonial contract consisted of a little over 100 marks (Shetland documents 1580-1611, No 419).

Compared to Scandinavian late medieval estates, and using the suggested number of c. 70 average tenant farms, Herdis’ estate would have placed her on the verge of high nobility in Denmark (80-100 farms), while her estate would have outdwarfed the combined estates of a Danish married couple of good standing from the lower nobility, which on average controlled c. 50 tenancies (numbers from Ulsig 1968).

Comparative material from Norway puts Herdis Thorvaldsdatter in the very top level among landowners. The Norwegian estate (that is, excluding the Shetland lands) of Hans Sigurdsson (Sigurd Jonsson) consisted of some 600 different properties, large and small (Sandberg 1970:182), compared to the plus 180 Shetland properties reconstructed here. The rather scarce description in the deed of inheritance of 1490 for the Norwegian estate, mirrors the one for the Shetland lands in the same document. The only surviving manorial cadaster from fifteenth century Norway, that of Hartvig Krummedike, indicates that the accumulated estate at the time of his death in 1476 consisted of 240 larger and smaller properties, which, however, would only have put him in seventeenth place among the largest contemporary estates in Denmark (Benedictow 1970:31-32, 34). While it has been hypothesized that the Giske estate might have consisted of as many as 200 farms in the late thirteenth century, lady Herdis’ estate compares well with two other estates, namely the Bjarkøy estate, which probably consisted of 18-20 farms in the late medieval period, and the Sudrheim estate, which was c. 80 farms (Holmsen 1980:32, 37 45).

Origins and organization of the estate
If we take a closer look at the three parts of the Shetland estate discussed in the following, an interesting pattern emerges. As pointed out by Barbara Crawford, ‘each family had an island which seems to have functioned as its estate centre and to which other scattered estates on the Mainland of Shetland, or in the other islands were attached’ (Crawford & Ballin Smith 1999:31). The islands in question are Papa Stour in Sandness, Vaila in Walls, and Noss, itself a small island lying off the larger island of Bressay.

All three islands seem to have belonged to the estate in their entirety, but otherwise they do not really compare. The fertile Papa, valued at 216 marks, was obviously much better placed to function as an estate centre in the traditional, manor-wise sense of the term, than the smaller, but still relatively significant Noss (60 marks), while Vaila with its 27 marks was relatively minor, and not even the largest of the Vaila properties in the parish of Walls.

On the other hand, Crawford’s suggestion that scattered estates around Shetland were attached to these three centres, seem to imply too loose a system. While stray properties were indeed a significant part of all three estates, there is a rather clear, and somewhat different, structure to them. Let us first look at the estate centred on Papa (for the following, see table 2).

216 marks of land, 46 percent of the known lands of the Papa estate, was situated on Papa Stour itself. The rest of the known Papa estate, 254 marks, however, was not really stray properties found around the islands, but were located in two clusters in Northmavine; one, totalling a little over 120 marks, in Eshaness on the western coast, facing Papa Stour, and another one, valued at 70 marks, clustered in and around the scattald of Ollaberry on the east coast. All together, the Papa ‘centre’ controlled 36 properties in Northmavine, 22 of which were situated in these two limited areas. Of the 16 properties in Northmavine which constitute 50 per cent or more of the farms/townships where they are located, no less than 14 are situated in Eshaness and Ollaberry, seven in each of them.

So, if we include Papa itself, the estate comprised three large and separate ‘blocks’ of land, with a relatively insignificant number of smaller pieces of property scattered in neighbouring areas. This, we might add, seems to be a similar structure to the one often characterising the larger estates in Norway in the late medieval period.

As for Noss, a rather similar, but not as clear structure emerges. Of the 370 marks belonging to the Noss estate, 120 (including Noss itself at 60) are situated in townships in Bressay, where 34 per cent of the lands belonged to the estate. Another 53,5 marks were situated in the neighbouring district of Tingwall. The rest of the properties are spread out over the eastern and north-eastern districts of Mainland, with 65,5 more marks situated in the districts immediately to the north of Tingwall, that is in Nesting, Lunnasting and Whalsay. All these areas are readily accessible from Noss by sea. To the south of Tingwall, a rather extensive area of the Mainland, the Noss estate owned only four, all of them minor, properties. There is as good as no overlap in the geographical distribution of lands belonging to the Noss estate and either of the other two estates. Even if the cluster of holdings in Bressay and Noss cannot really compare with Papa, it still represents a definite core area within an estate that is much more compressed than the impression one gets from the available literature.

What about ‘Vaila and Vailaguids’, then? To begin with, it is not as localised as the other two estates. Instead, there are two distinct distribution areas; the western part of the Mainland (Walls, Sandness) and the islands in the northeast (Unst, Fetlar). The map (figure 1) shows distinct clusters of major holdings in the Haroldswick area in Unst and in Walls. These look like core areas similar to the ones in Papa, Northmavine and Bressay. In a way they are, but the picture still looks very different from what we just discussed regarding Papa and Noss. The reason for this, is that the deeds for many of the properties in just these two core areas have survived. As we have seen above, Herdis Thorvaldsdatter bought lands on two separate occasions – in 1355 and again in 1360. These two transactions added some 150 marks of land to her estate.

The first acquisition gave her Footabrough and three other properties in Walls. The Footabrough property (36 marks) purchased in 1355 was the biggest of lady Herdis’ lands in Walls, with only Vaila and Foula approaching a similar size. If we exclude the properties bought in 1355, Walls does not look much like a core area. Similarly, for Unst, the transaction of 1360 is the only reason for the cluster of lands in Haroldswick. The beyond comparison largest property belonging to the Vaila estate in Unst, the island of Uyea, was also part of the same transaction. Without these lands in Unst and the ones in Walls, the Vaila estate comes forward first and foremost as a collection of smaller pieces of properties widely distributed across Shetland. There are still a few larger properties, and these are found mainly in Sandness and Walls, but there is no’core’ of clustered, big properties comparable to the ones we find in Papa Stour and, to a lesser degree, in Noss/Bressay. Without the lands added in 1355 and 1360, the size of the Vaila estate would be on par with the Noss estate.

Another document of 1355 seems to imply that Herdis was selling lands, as well (Shetland documents 1195-1579, No 10; cf DN I, 340). In Shetland, as in the Norwegian kingdom at large, the udal lands were not inalienable, as it was for instance in Denmark, where changes in the composition of aristocratic estates were almost exclusively due to intra-aristocratic transactions (Benedictow 1970:13).

The two deeds of 1355/1360 are the only ones we have got for the extensive estate controlled by lady Herdis. Perhaps it is significant that both the transactions that we know her to have been involved in, deals with the Vaila part of the estate. It is possible that this signifies a different status for this part of the estate than for the other two. But is the tripartite division that is documented in 1490, relevant for discussing the situation in Herdis’ own lifetime?

Chances are that it is. The formulaic ‘Papa and Papa gods’ etc is the common way of denoting manor farms (Norw. setegårder) and the lands belonging to them in the Norwegian language in the time period in question (fourteenth to sixteenth centuries). Furthermore, there are many examples of such estates being considered and managed as discrete entities even if they were swopped up by bigger landowners with more extensive estates. Even if the Æri estate in Western Norway was passed on to two successive families in Eastern Norway and remained in their possession as a part of two different (and larger) estates for two or three generations in the late thirteenth and early to middle part of the fourteenth century, it was managed as a discrete estate during the whole period and always called ‘ Æri gods’. Two years after the Reformation in 1537, Truid Ulfstand on behalf of his wife Gørvel Fadersdatter demanded that ’Nausdall og Nausdals gotzs’ should be returned to her from the cathedral in Bergen (DN IV, 1115). A similar demand was raised for ‘Torgar og Torgar gods’ in Northern Norway, c. 1533. The latter was not granted, since it was found that it had been owned by seven successive archbishops (in Trondheim), without anyone else claiming ownership over it (DN IV, 1101; see also Iversen 1999:58).

Thus, it would not be at all surprising if the division documented in the deed of inheritance of 1490 reflects the situation in the mid-fourteenth century, even if the (suggested) three separate estates had then been part of the bigger Sudrheim – Giske estate for more than one hundred years. The most likely reason why the Norwegian lesser estates often were kept as such, even if they were bought or inherited into much larger estates, seems to us to be that they were regularly organised as estates, too, with a central manor surrounded by smaller, dependent farms and with other attached properties at a greater distance. Even if it was just for collecting the yearly rent, this might – in a time and place with relatively modest means of communication – be looked upon as the best way of structuring much bigger estates, as well, thus leading to what are in effect multipolar estates.

We suggest, then that lady Herdis’ Shetland estate was a multipolar one, and that it was comprised of three, discrete estates, probably with different origins, and each perhaps having once been independent holdings. The Papa estate has a structure that might point to this being a multipolar estate in itself. As for what kind of origins the three estates had, it can hardly be more than an educated guess. It is certain that Papa Stour was controlled by the Norwegian king at the end of the thirteenth century, and that it was passed on to Thorvald Thoresson or one of his children in the early part of the fourteenth. Many noble families in the service of the king expanded their estates through the money and position they got as lendmenn or sysselmenn (Skre 1998:62).

Crawford’s hypothesis, that the Papa estate belonged to the Orkney earls before Shetland was put under immediate royal control in 1195, seems a distinct possibility (Crawford 2002:20). But one should not forget that much of the lands that the Norwegian king gave away as gifts or redisposed of in the thirteenth and fourteenth centuries, were properties that only recently had come under royal control (Bjørkvik 1968:133). The Noss estate might have a similar background, and it is possible that it was managed alongside the Papa estate for a considerable period of time before it got into lady Herdis’ hands. The similarities between these two estates and many of the royal and aristocratic estates found in Western Norway when it comes to structure and situation on islands at or near major sea routes, are interesting, and could be a pattern with roots in the Viking Age (cf Bjørkvik 1958).

The Vaila estate seems to have a different background. It hardly approaches the estate model found in Papa and Noss, or the examples we have mentioned from Norway. This might be lady Herdis’ own lands, perhaps put together piece by piece through a combination of inherited stray properties, gifts from her husband(s), and bought lands. What little overlap there is between the three estates in the lady’s possession, three instances, all involves the Vaila estate (Laxo in Delting, Frangod and perhaps Priesthoulland in Northmavine). This might suggest that there has been a certain degree of adjustment between the Vaila estate and more peripheral areas of the other two.

References
NAS=National archives of Scotland.
Benedictow, O. J. 1970. Hartvig Krummedikes jordegods. En studie i en senmiddelaldersk adelsøkonomi. Oslo.
Bjørkvik, H. 1958. Jord-eige og jord-leige i Ryfylke i eldre tid. Fordelingen av jordeigedomen på Vestlandet og Agdesida i 1661 og i seinkatolsk tid. Stavanger.
Bjørkvik, H. 1968. Det norske krongodset i mellomalderen. Unpublished manuscript.
Crawford, B. E. 1983. The pledging of the islands in 1469; the historical background. In D. J. Withrington (ed), Shetland and the outside world 1469-1969, pp 32-48. Oxford.
Crawford, B. E. 2002. The historical and archaeological background to the Papa Stour project. In B. E. Crawford & J. G. Jørgensen (eds), Papa Stour and 1299, pp 13-35. Lerwick.
Crawford, B. E. & B. Ballin Smith 1999. The Biggings, Papa Stour, Shetland. The history and excavation of a royal Norwegian farm. Edinburgh.
Daae, L. 1895. Om Fru Gørvel Fadersdatter og hendes norske Jordegods. Historisk Tidsskrift 13, pp 219-274. Kristiania.
DN=Diplomatarium norvegicum. Christiania/Oslo, 1849-..
Holmsen, A. 1980. Gård og gods i Norge i eldre tid. Oslo.
Islandske annaler indtil 1578. Published by Gustav Storm, Christiania, 1888.
Iversen, F. 1999. Var middelalderens lendmannsgårder kjerner i eldre godssamlinger? En analyse av romlig organisering av graver og eiendomsstruktur i Hordaland og Sogn og Fjordane. Bergen
Sandberg, P.-Ø. 1970. Jordeboka over Hans Sigurdssons norske godskompleks. Et forsøk på rekonstruksjon av Sudrheimsættens jordegods i Norge ca. 1500. Thesis, Oslo University.
Shetland documents 1580-1611. Edited by J. H. Ballantyne & B. Smith. Lerwick, 1994.
Shetland documents 1195-1579. Edited by J. H. Ballantyne & B. Smith. Lerwick, 1999.
Skre, D. 1998. Herredømmet. Bosetning og besittelse på Romerike 200-1350e.Kr. Oslo.
Smith, B. 1990. Shetland, Scandinavia, Scotland, 1300-1700. In G. G. Simpson (ed), Scotland and Scandinavia 1300-1800. Edinburgh.
Smith, B. 2000. Toons and tenants. Settlement and society in Shetland, 1299-1899. Lerwick.
Sollied, H. 1942. Kildekritiske undersøkelser vedrørende nogen middelalderslekter. Norsk slektshistorisk tidsskrift, vol VIII, pp 126-129. Oslo.
Ulsig, E. 1968. Danske adelsgodser i middelalderen. Copenhagen.
Øye, I. 2002. Ragnhild Simonsdatter and women’s social and economic position in Norse society. In B. E. Crawford & J. G. Jørgensen (eds), Papa Stour and 1299, pp 81-93. Lerwick.

03 mars 2007

"Glædelig jul og godt nyt år" fra A. W. Brøgger

Den må ha gjort inntrykk på A. W. Brøgger, den merkelige gruppen med flatmarksgraver han undersøkte på Bringsvær i Fjære ved Grimstad senhøstes 1922. Iallfall var det den ene graven, komplett med skjelett og det hele, som prydet forsiden på de julekortene han sendte til kolleger i inn- og utland senere samme år. "Glædelig jul og godt nyt år" står det på baksiden av kortet, hvis forside er gjengitt her.

Hva var det da som fascinerte den rutinerte arkeologen? Rike gjenstandsfunn kan det ikke ha vært, selv om den avbildede graven inneholdt gravgods i form av en skålvekt, en skinnpung med seks vektlodd, en tysk sølvmynt og en plommekjerne, en kam, en kniv og en øks. De to øvrige gravene var imidlertid tomme. Nei, det var den spesielle gravskikken Brøgger hadde festet seg ved. For de tre gravene som ble påtruffet i forbindelse med nydyrkning på den gamle storgården Bringsvær, en drøy kilometer fra Fjære kirke, skiller seg på vesentlige måter fra vikingtidsgraver ellers i det sørlige Norge.

I de sørnorske kystdistriktene ser det ut til at den hedenske gravskikken i all hovedsak forlates omkring 950 eller tidligere. Kun en håndfull graver på Agderkysten kan dateres til etter 950. Det dreier seg først og fremst om de tre fra Bringsvær. Mynten i den ene gravlagtes pung daterer graven til etter 983. Det i seg selv gjør denne gravgruppen spesiell.

Men også gravformen er bemerkelsesverdig, idet samtlige tre gravlagte var lagt i kister eller kamre av stein. Steinkisten som kan minne om disse, hadde vært vanlig blant aristokratiet i yngre romertid og folkevandringstid i denne landsdelen, men det er ikke spor etter skikken etter begynnelsen av 500-årene. Hadde det ikke vært for at gravgodset i den ene graven så utvilsomt peker mot sen vikingtid, ville trolig Brøgger ha ment at det dreide seg om graver fra folkevandringstid. Og selv om det ikke er helt uhørt med steinkister under flatmark i folkevandringstid, så er datidens steinkistegraver nesten alltid dekket av en gravhaug. På Bringsvær var det ikke spor etter gravhaug over noen av de tre gravene.

Dessuten var kistenes orientering uvanlig. Det klart vanligste i jordfestegraver fra vikingtiden i Sør-Norge er at graven er orientert nord-sør, med hodet i nord. Det finnes bare et fåtall graver med avvikende orientering. I Danmark er derimot Ø-V-orientering vanlig i jordfestegravene. Bringsværgravene var orientert mer eller mindre NØ-SV.

Den største kisten inneholdt viste seg å være bygd for en mann i 30-årsalderen. I graven fantes foruten et så å si intakt skjelett, gravgods som nevnt ovenfor. Plasseringen av gjenstandene i graven tyder på at alle sakene – med unntak av øksen – har vært opphengt i mannens belte. De to andre kistene var mindre, og den ene inneholdt restene av et barneskjelett.

Siden Brøggers tid har man diskutert om gravskikken i Bringsværgravene skal regnes for hedensk eller kristen. Steinkister som minner om dem fra Bringsvær, er kjent fra kristne gravplasser i tidlig middelalder – først og fremst fra De britiske øyer, men også fra hjemlige funn som den antatte kirkegården på Tanberg på Ringerike i Buskerud . Av utstyret i mannsgraven er det bare øksen som ikke direkte kan sies å ha vært del av det personlige utstyret. Spørsmålet om Bringsværmannen var kristen eller ikke, kan ikke avgjøres uten nye arkeologiske undersøkelser på funnstedet. Men ut fra de tilgjengelige funnopplysningene kan man ikke helt avvise den tanken at de særpregede gravene tilhører en tidlig kristen gravplass eller endog en kirkegård.

I grenseland

Hvordan har gårdene rundt i landet fått grensene sine? Dagens grenser kan ha en skiftende historie bak seg – ofte preget av strid og uenighe...