CLAN CARRUTHERS INT SOCIETY CCIS

THE HISTORY OF GUTLAND

The Discovery of Gotland

Sinking Island

Hallowing with Fire

Mythical or Mystical Ancestors

The Settlement of the Island

Dreams About Snakes

Predictive Verse

The Division of the Island

Emigration as a remedy for over population

Torsburgen and Faro

Tricking of the King

Heathen Beliefs and Practices

Treaty with Sweden

The story of the discovery of Gotland and the name Þieluar pose
the first problem: to determine who Þieluar might have been, and
what historical or traditional connection, if any, he had with Gotland.
A variant of the name is known from a pair of runic inscriptions
found on stones in Öster Skam in Östergötland; cf. Note to 2/1.
These runestones presumably predate the original text of Guta
saga, although according to Brate (ÖR, 25–27) doubt has been
placed on their antiquity, and it was suspected that they were the
work of a seventeenth-century antiquarian, although Brate himself
considers the inscriptions as recorded to be genuine. P. A. Säve
writes (1862, 59) that he was unable to find evidence of the stone
or stones and that no one in the parish could offer any information,

even the local dean, who was of the opinion that they did not exist
and had never existed. There might seem to be no great similarity
between the Þieluar of Guta saga and Þjálfi, Thor’s servant in
Snorri Sturluson’s Edda (1982, 37, 40, 43, 177), although it has
been suggested by several scholars, including Läffler (1908–1909,
Part 1, 170–171) and Gordon (1962, 255), and has been taken up
by Uwe Lemke (1986, 14), who sees Þieluar as the representative
of Thor, the thunder, lightning and life god. In that role, he would
be an appropriate agent to free the island of its enchantment. Olrik
(1905, 136–138) wonders if there is not a mystical aspect to Þieluar/
Þjálfi, despite the fact that he is usually in Thor’s shadow, but
points out that fire is so commonly called upon to dispel spirits that
the world of the gods need not be involved. Olrik notes also that the
name is thought to be related to a presumed Icelandic *þjálf,
‘work’, leading to speculation that he might be a work-god, but
dismisses much of this as mere conjecture. No such form occurs in
Old Icelandic, but it is worth noting that Þjálfi could be a weak
form of Þieluar; cf. ÍO, s. v. þjálfa; de Vries, 1956–1957, II, 129.
The existence of oral traditions in connection with Þieluar is
perhaps shown by the fact that a later Bronze Age grave (1000–300
BC), near the east coast of Gotland and lying almost directly east of
Visby, in the parish of Boge, is called Tjelvars grav. This is not,
however, the oldest grave in Gotland, which was inhabited prior to
the Bronze Age, and even if it were, the name could be secondary
to Guta saga, so it cannot be regarded as a source.

B Sinking islands
In relation to the legend of the island of Gotland sinking by day and
rising up by night, there is geological evidence to support a number
of changes in sea level and these could well have been compressed
in folk memory into a diurnal change, followed by the final fixing
of Gotland above sea level. Gotland was below sea level at the end
of the last Ice Age, having been above it previously, and it appears
that the sea level then slowly fell, resulting in a series of steps in
the coastline; cf. Klintberg, 1909, 33, 35–36. What is certain,
geologically, is that the sea was once very much higher than it is
today and there could therefore have been a period during which
parts of the island at least were sometimes above sea level and
sometimes below it. The various levels of sea-wall testify to this
and geologists point to the movements in the Baltic basin and the

sinking of the land in the Ice Age as a cause; cf. Lemke, 1970, 4.
Gotland itself is relatively flat, so if the sea level were near to the
top of the present cliffs, it might well seem as if the island were
disappearing and re-appearing in a mysterious manner, especially
in bad weather. That being the case, it is not surprising that some
folk memory remained of this period and that it was included in a
legendary history of the island.
There are, however, legends from Iceland and other parts of
Scandinavia, Ireland, Finland and England, which can be cited as
related to the motif of a floating island. Particularly fertile islands
were frequently the subject of legends concerning their magical
origins; cf. Gordon, 1962, 255–256. Several islands are deemed to
have been disenchanted by fire or steel. Svínoy, the most easterly
of the Faeroe Islands, is one of a number mentioned by Strömbäck
(1970, 146–148) in this connection. Svínoy is described by Lucas
Debes (1673, 21–22) as a flydøe bewitched by the devil, which had
to be ‘fixed’ with steel. A similar tale from 1676 is told of the
mythical island Utröst, west of Lofoten in Nordland, and from later
times of Sandflesa (west of Træna), Utvega (to the west of Vega),
Hillerei-øi, Ytter-Sklinna (in Nord-Trøndelag), and other islands in
Norway; cf. Storm, 1895, 208; Nansen 1911, 286. Steel would have
rendered them visible and thus disenchanted, but they were too far
out to sea to have been reached by a domestic animal needed to
carry it, so they remained submerged. The ‘lucky’ island O’Brasil
or Hy Breasail, off the west coast of Ireland, was said only to appear
every seven years and would stay in sight if someone could throw
fire on it; cf. Nansen 1911, 287. Giraldus Cambrensis (1867, Part 2,
94–95) writes in the twelfth century of an island off the coast of
Ireland, which disappeared as a group of young men attempted to
disembark, but was ‘fixed’ by an arrow of red-hot iron being
thrown on it as they approached. Another island off the west coast
of Ireland, Inishbofin, ‘white cow island’, was ‘fixed’ when two
sailors landed on it and lit a fire; cf. Palmenfelt, 1979, 128. William
of Malmesbury (1981, 44–47, 52–53) tells the story of Glasteing
being led by his sow to the island of Avalon at Glastonbury and
there are tales of islands, including Svínoy, being disenchanted by
tying steel to a sow that was in the habit of visiting the island.
Whether any of these tales could have been known to the author of
Guta saga is difficult to assess, although such stories were clearly
common, at least throughout Europe. Spegel (1901, 22) lists several more of them in his account of the history and geography of
Gotland. First, the island of Delos in the Aegean, which was said
by the Greek poet Callimachus, in his Hymn to Delos (c.275–262
BC), to have sailed to and fro over the sea, being sometimes visible
and sometimes not; see Mineur, 1984, 75–95. In legend, the mysterious Delos was said to have been called from the bottom of the
sea by Poseidon and eventually chained to the sea-bed by Zeus.
The island was the birthplace of Apollo and Artemis and sacred to
the former. Secondly, Spegel cites the island of San Borondon, of
which the sixteenth-century Dutchman, van Linschoten, reports
that the Spaniards thought it lay about 100 miles west of the
Canaries; see Linschoten, 1598, 177. They could see the island, but
never find it, and assumed it was either enchanted, or small and
covered with clouds. Thirdly, Chemmis, an island on a lake at the
mouth of the Nile, which Herodotus (Book 2, §156) was told
floated while it was being used as a hiding-place for Apollo. Cf.
also Nansen, 1911, 283–285 and references.

C Hallowing with fire
The motif of hallowing or removing a spell with fire is found
widely in Scandinavia and there is also evidence for its actual
occurrence. Examples are found in Danish, Icelandic and Irish
literature, and have been discussed in detail by Strömbäck (1970,
142–159). He supports the theory that as well as, or as an alternative
to, any legal implications, the ringing of land with fire in some way
placated the land spirits who had bewitched it. To reinforce this he
interprets eluist as a form of eluiskt meaning ‘bewitched’, and
suggests that the account in Guta saga represents merely a more
pointed version of the belief lying behind land-claiming customs,
similar to those mentioned concerning Jo≈rundr goði in Landnámabók
(ÍF I, 350, 351) and Þórólfr in Eyrbyggja saga (ch. 4; ÍF IV, 8) as
well as in Vatnsdœla saga (ch. 10; ÍF VIII, 28) and Hœnsa-Þóris
saga (ch. 9; ÍF III, 25). These are, however, simply parallels and it
is probable that there was a similar oral tradition associated with
Gotland itself. The idea lying behind the fire legend may be that the
island could only be inhabited once it was dry enough to sustain
fire, that is, when the water level was low enough.

D Mythical or mystical ancestors
Having removed the spell from the island, Þieluar disappears from
the scene and is not mentioned as a permanent settler. This puts
him in the role of ‘mystical ancestor’, on a parallel with Tuisto,
referred to by Tacitus (1914, 32), and the Gothic Gaut/Gapt, in
Jordanes (1997, 70), discussed, along with other examples, by
Schütte (1907, 135–136). There are many parallels for such an ancestor and it is not possible to determine whether one of the accounts
extant at the time was the inspiration for the tale incorporated by
the writer of Guta saga, or whether a separate oral tradition existed.

E The settlement of the island
There follows the description of a further two generations of mythical
ancestors from whom the inhabitants of Gotland are deemed to
have descended. Hafþi, the son of Þieluar, marries Huitastierna
(‘white star’) and they have three sons who are all given names
starting with ‘G’. Legendary genealogies consisting of sets of
alliterating names are common in the early histories of several
peoples. According to Tacitus (1914, 32), the Germans worshipped
an earth-born god, Tuisto, who had a son Mannus, himself the
father of three sons, the founders of the three races of Ingaevones,
(H)erminones and Istaevones, who were celebrated in songs. These
sons were named Inguo, Ermenus and Istio in sixth-century sources;
cf. Tacitus, 1914, 136. Similarly, the Gothic tribes recognised one
ancestor by the name of Gapt (or Gaut), who had a son named
Hulmul (Humli or Humal), called the father of the Danes, himself
the father of Augis (Agis or Avigis), the father of Amal, the father
of Hisarna, the father of Ostrogotha, the father of Hunuil, the
father of Athal and so on; see Jordanes, 1997, 70; Wolfram, 1988,
31. In medieval Scandinavian literature, several mythical genealogies
are mentioned, including those in Snorri’s Gylfaginning and Ynglinga
saga. In Gylfaginning (Snorri Sturluson, 1982, 11) Snorri writes of
Auðhumla, the giant cow, which licks Búri out of a block of salt.
This Búri marries Bestla and has a son Borr, in turn the father of
Óðinn, Vili and Vé. The earliest extant version of a genealogy of
the Norse people is probably represented by Upphaf allra frásagna
(ÍF XXXV, 39–40), which is thought to be the beginning of a lost
Skjo≈ldunga saga. In it, Fróði, the great-grandson of Óðinn, is described as a bringer of peace and prosperity and a contemporary of
Christ; cf. Faulkes, 1978–1979, 94–95, 107–108. In Ynglinga saga
ch. 10–13 (ÍF XXVI, 23–29) Snorri gives the genealogy of Yngvifreyr’s line and in ch. 17 (ÍF XXVI, 34) that of Rígr father to Danpr,
grandfather to Drótt and great-grandfather to Dyggvi. Again, in
Landnámabók (ÍF I, 40) the three sons of Atli are Hásteinn, Hersteinn
and Hólmsteinn, although these could be historical. Keil (1931,
60–70) suggests that the choice of names in the Icelandic sagas
was also partly influenced by alliteration and other similar factors.
Considering the specific names in Guta saga, the name Hafþi
might possibly be linked with the parish name Havdhem on southern
Gotland as Wessén (SL IV, 302) implies. It is more than likely,
however, that the parish name preceded the writing of that portion
of Guta saga, and that the name Hafþi is secondary to that. It is
necessary to accept the possibility that personal names appearing
in legends could have been invented as a result of the existence of
place-names with an apparent genitive form and/or with a second
element that invited such an assumption. Another example of this
possibility is Lickershamn, discussed below, p. xliv. Olsson (1984,
26) interprets Havdhem as relating either to the Gotlandic haued,
‘head’, or to havde, ‘raised grass bank at haymaking’. Schütte
(1907, 136), however, relates the name Hafþi itself to ‘head’,
suggesting he was the ‘head-man’, with the mystical wife, Huitastierna.
The place-name Havdhem, although it could have suggested the
name Hafþi, is not mentioned in Guta saga. The name Huitastierna,
apart from alliterating with Hafþi, leads Läffler (1908–1909, Part 1,
171–172) to note that it reminds one of the ‘cow-name’, and to
consider that the two might originate in an alternative creation myth,
representing animal deities; cf. above, p. xix, in relation to Inishbofin.

F Dreams about snakes
The dream that Huitastierna has on her wedding night, of the three
snakes issuing from her womb or breast, has folklore parallels. The
motif of pregnant women dreaming of events connected with the
birth of their children is very commonplace. There is, for example,
a tale concerning William the Conqueror’s mother who is said to
have dreamed that a great tree grew from her womb. Equally,
dreams concerning snakes are not unusual and the combination of
the two motifs (with the snakes proceeding from some part of a
woman’s anatomy) is also encountered. Henning Feilberg (1886–
1914, IV, 316, s. v. orm) mentions a motif concerning a snake

growing out of a young girl’s back and coiling itself around her
neck. Snakes also figure largely in Celtic myth in various guises:
as protectors, as fertility symbols and in connection with the underworld and death.
The snake motif is common on Gotlandic picture-stones and one
in particular, from Smiss in Gotland, is of interest; see Note to 2/8.
It is therefore possible that a literary or oral motif concerning a
pregnant woman’s dream has been combined with snake iconography to give this version of the tradition. What the true source is
for the dream-sequence it is probably not possible to know: it could
have been a folk-tale applied in a particular case or it could have
been a specific story associated with the island’s settlement, perhaps linked to some native or foreign mythological element. It
could even have been an invented story based on the seeds of an
idea sown by some artefact similar to the disc found in a woman’s
grave at Ihre, Gotland; cf.

Ardre Chapel

Ardre Chapekl
G Predictive verse
Huitastierna tells her husband the dream and he interprets it, by
means of a verse. The verse is delivered in two half-strophes, each
of three lines. The first half-strophe is a confirmation of the power
of fate, a reassurance and a statement of belief in the future. The
verse is in all probability older than Guta saga itself, i. e. not the
work of the author of Guta saga, and thus possibly the kernel of an
oral tale. In the second half-strophe, Hafþi gives his offspring
names ‘unborn as they are’: Guti, Graipr and Gunfiaun, and indicates that they will be born in that order, with the first taking the
lead in ruling Gotland. The place-names Gute (in the parish of
Bäl), Gothem, Gothemhammar, Gothemån, as well as Gotland
itself, would be apparently explained by this tale; cf. Note to 2/1.
There are no major place-names that obviously relate to the names
of the two other sons, and indeed it has been maintained by Wessén
(SL IV, 302) amongst others that the name Gunfiaun is unknown
outside Guta saga, although the element Gun- occurs in many
Scandinavian names. Schütte (1907, 194) suggests that it may be
a name plucked out of the air to complete the expected trinity of
names, and that the whole episode expresses a parochial view of
events. The name Graipr occurs, but only rarely, in Old Norse
literature. In the parish of Garde, however, there are remains said
to be of ‘Graipr’s house’ and ‘Graipr’s grave-mound’ (rör). There

is also a ruin in the parish of Ardre with the name ‘Gunnfiaun’s
chapel’, which is probably from the fourteenth century and therefore, like the other remains, very possibly secondary to the legend,
if not to Guta saga itself; cf. SL IV, 302–303. As Schütte (1907,
194) also points out, the three alliterating names must in any event
be regarded as a pure fiction, on the pattern of the three sons of
Borr, the three sons of Mannus and other examples. In Guta saga,
the names could be being used merely as an explanation for the
division of the island into thirds. In fact what immediately follows
is a contradiction of the verse just quoted, a not uncommon phenomenon in Old Norse literature when an older verse is incorporated into a prose work by a later author. In the verse, Guti is
presented as the first and most important son, who will own all
Gotland, whereas the prose following cites Graipr as the eldest
son, with Guti taking the central position. The fact that the middle
third of Gotland contains Roma, later the site of gutnalþing, the
assembly for the whole island, could have influenced this version
of events, and show that it might represent a later tradition in which
Guti was associated with the middle third; cf. Notes to 2/19; 2/27;
6/21–22. In this case, as Wessén suggests (SL IV, 302–303), the
only ‘error’ in the prose text of Guta saga is in the naming of
Graipr as the oldest son.
The strophe with which Hafþi interprets his wife’s dream may
well be part of a longer poem and the alliteration in the prose
surrounding the verse (e. g. sum hit Hafþi, sum þaun saman suafu,
droymdi henni draumbr, slungnir saman, skiptu siþan Gutlandi,
lutaþu þair bort af landi) suggests that the material in it appeared
in the lost verse; cf. Notes to 2/12–14, 16–18. Lindquist (1941, 12,
39, 51) has discussed in detail the lists of bishops and lawmen that
appear as supplements to Västgötalagen and argues convincingly,
by only slightly rewording the prose, that they are the remnants of
now lost verses. It is possible that a similar literary source lies
behind at least this early part of Guta saga. If there were two
versions extant, differing slightly in the tradition they represented,
this would explain the apparent contradiction in the text.

H The division of the island
One possible explanation for the discrepancy between the verse
and the prose describing the division of the island has been given
above. Läffler (1908–1909, Part 1, 172–177), on the other hand,
argues that the verse carries a separate tradition from that behind
the prose, and that it has been included here in order to follow
Saxo’s example of larding his texts with verse. Authors such as
Strelow (whose chronicle of Gotland, Cronica Guthilandorum,
was designed to show that the island had been settled by, and
always subject to, the Danes, in particular the Jutar) have even
more imaginative ideas, based on surmise from place-names; see
Strelow, 1633, 20–21. Strelow, in fact, names the two younger sons
as Grippa and Gumphinus. There may have been several versions
of an oral genealogy and associated stories of the division of the
island, but the straightforward explanation, given above, is preferable to a more complicated one. The connection between the three
sons of Hafþi and the three administrative districts had certainly
been made by the time the verse source was composed.
The division of the island of Gotland into three is first recorded
in 1213 (DS I, 178, no. 152) in a letter from the Pope to the deans
of the northern and southern thirds (‘prepositis de Northlanda et de
Sutherlanda’) and to the abbot of Gotland (‘abbati de Gothlanda’),
who would have been in spiritual charge of the middle third; see
Yrwing, 1978, 81. A recent study by Hyenstrand (1989) questions
the age of the þriþiung division, however, and argues that the
original division of Gotland was not into thirds. He suggests that
the original division was into 12 hundari and that this was older
than that into þriþiungar, although it is only mentioned in Guta lag
and not elsewhere; cf. GLGS, 46. He notes that the number 12
appears frequently in Guta saga, and suggests that the original
administrative division of the island was into 12 hundari, each
divided into eight, which later gave rise to the division of the
island into sixths, settingar (Gutnish pl. siettungar), and thirds; see
Hyenstrand, 1989, 108, 136 and cf. Note to 2/19. Both hundaris þing
and siettungs þing are referred to in Guta lag.
The motif of the division of an island, in this case Ireland, occurs
in Giraldus Cambrensis (1867, Part 3, 143–147), where successions
of brothers, some with alliterating names, divide up the land between them, before one becomes king of the whole of Ireland.

I Emigration as a remedy for over-population
The enforced emigration resulting from the overpopulation of Gotland
may have been a historical reality and possibly the subject of oral
tradition, as is discussed below. On the other hand there are so
xxvi Guta saga
many instances of similar events in the history of various peoples
that the possibility that the author was adapting a literary motif
must also be considered. Weibull thinks that this whole episode is
a formula tale, ‘en lärd transponering på ett nytt folk av en urgammal
utvandringsberättelse’; see Weibull, 1963, 27, 34–35. He considers
it to be derived ultimately from authors such as Herodotus (Book 1,
§94), in which the author writes of the King of Lydia dividing the
people into two groups, determining by lot which group should
emigrate and which stay at home. Weibull argues that, as the tale
appears so frequently in other sources, it cannot be true in the
particular case of Gotland. A similar story, indeed, occurs in Book
8 of Saxo’s Gesta Danorum (written around 1200), with the added
twist that the original suggestion was to kill the old and the very
young and to send away those below arms-bearing age; cf. Saxo
Grammaticus, 1931–1957, I, 237–238. By general agreement this
plan is rejected and there is a ballot, after which it is the stronger
members of the community who must stand in for the weaker
members in exile. They set off from Denmark, stopping in Blekinge
and, coincidentally, anchor off Gotland on their way eastwards.
They are instructed by divine intervention to change their name to
Langobardi and eventually reach Italy, where they impose their
name upon the existing inhabitants. Saxo refers directly to Paulus
Diaconus’s history of the Langobardi (written at the end of the
eighth century), where there is a similar account, including mention
of Nigilanda or Ingolanda (or ‘Golanda’, derived from Golhaida)
as one of the places visited on the way to Italy; cf. Paulus Diaconus,
1878, 54. There is another version of the motif, but this time relating
to the young men of Dacia under Rollo, in Dudo of St Quentin’s
History of the Normans, written about 1014 (Part 2, §1–2, 5).
There are certainly considerable similarities between these stories,
of which the one in Guta saga was most probably the latest in
written form, but this does not necessarily mean that its writer
consciously borrowed from those cited, or from any similar source.
If, on the other hand, the possibility of the episode recording
details of an actual exodus is accepted, the question of when that
exodus occurred has to be considered. One wonders if over-population
might not, in some areas, have been a cause of Viking activity,
although archaeological studies have led scholars to date the exodus
from Gotland, certainly, several centuries earlier than the Viking
Age. The archaeological evidence points to a sharp reduction in
The history of the Gotlanders xxvii
population between circa 475 and 550, as indicated by the paucity
of grave finds and by the number of abandoned settlements; cf.
Nerman, 1963, 19. At the same time, the instances of imported
goods from Gotland increased in the countries around the eastern
Baltic; cf. KL, s. v. Vikingetog, cols 49–51.
If the exodus did occur, the story in Guta saga probably relates
to folk-tales generated from this period. It is possible that some
of the banished Gotlanders or their descendents came back years
later with exotic goods and tales of the East, but of this there is no
remaining direct evidence. From finds in Gotland, it appears that
during the ninth and tenth centuries the coins imported were principally from the Caliphate, and there are very few from Byzantium
itself. Later, coins seem to have come chiefly from western Europe.
If there was any group returning from exile, it does not appear to
have been a large one. Hadorph (1687, viii) considers the emigration
episode to be important in relation to the start of the great Scandinavian
expeditions, but this is not supported by the available evidence.
Nils Tiberg (1946, 44) suggests it would have been natural for
both the author of Guta saga and the composer of the material he
used to have had patterns in mind. Having received an oral tale that
he wanted to record, the author might expand it to some extent on
the basis of similar written material. The opposing argument in
favour of purely literary sources has been discussed, but it seems
probable that behind the tale presented here there is some genuine
oral material that relates specifically to Gotland. One point worth
noting in the story as told by Paulus Diaconus is that he writes, of
the island of Scandinavia (variously Scadinavia and Scadanavia),
that it is covered by the waves that run along its flat shores; see
Paulus Diaconus 1878, 48–49, 52, 54; Goffart, 1988, 385. This, or
some similar tale, is another possible source of inspiration for the
discovery legend. Cf. also Olaus Magnus, 1909–1951, I, Book IV,
ch. 6.

(Torsburgen Hill Fort on Flickr. Located near the east coast of Gotland, Torsburgen is the largest fortified hill-fort in Scandinavia, with an area of 112.5 hectares. It’s wall has a diametre of about a kilometer and some remains of it are up to 7 meters high. Built around the 1st century AD and reinforced during the 4th century, Torsburgen fort was in use until c. 1100 AD. A small portion of the wall has been restored after archaeological excavations.)

J Torsburgen and Fårö
According to the text, the people who were balloted away declined
in the end to depart and installed themselves in Torsburgen, called
Þorsborg in Guta saga. This immense prehistoric fortification, the
largest of Sweden’s hill-forts, utilises one of the few high places on
the island, so that man-made fortification was only required along
half its perimeter. Considerable archeological research has been

done into the dating and use of Torsburgen, and it is certainly
possible that it was used in the way suggested in Guta saga. It is
impressive in scale and could have supported several thousand
people. Engström (1984, abstract on title verso, 123, 124–126; GV,
76) has estimated that about 100–200 men could complete each
two-kilometre length of wall in approximately two months. The
date of construction has been disputed, but as a result of radiocarbon dating and other techniques, Engström has dated the two
phases of the fort to the periods between AD 300 and AD 400, the
end of the Roman Iron Age, and between AD 800 and AD 1100, the
end of the Viking Age. These were periods of vigorous Scandinavian
expansion, combined with social and climatic change, which might
have been the cause of unrest. The position of Torsburgen, near to
the coast and to administrative centres, would lend itself to use in
the defence of the island, as well as in any internal conflicts. The
suggestion (Engström, 1979, 127–128) is that Torsburgen was
constructed as a defensive fort from which the islanders sortied to
fall upon invaders. It is possible that this successful strategy lies
behind the later episode concerning the ‘many kings’ who attacked
Gotland, but the author does not mention Torsburgen specifically
in connection with these attacks. Engström rejects the suggestion
that Torsburgen was a general place of refuge, since it lay too near
the coast from which danger might come, but does not dismiss the
idea that it might have been built by a group of Gotlanders threatened with expulsion. The findings are, in general, consistent with
the emigration story as recorded in Guta saga, which gives the
clear impression that Torsburgen was established well before the
emigrants fled there, and was not built by them. There is also
evidence that at least one of the walls has been augmented after its
initial construction. However defensible their position was, the
emigrants were not permitted to stay at Torsburgen and decamped
to the island of Fårö, where they again failed to set up a permanent
residence; cf.
One might expect an edifice of the size and prominence of
Torsburgen to attract oral traditions, but it has a relatively low-key
role in the story as told in Guta saga. In the sixteenth- and
seventeenth-century chronicles of Gotland the forced emigrants
are led by one Tore, and Torsburgen is said in one tradition to have
been named after him; cf. Strelow, 1633, 32. It seems more likely,
however, that the name relates to a cult place dedicated to Thor and
that later authors have combined the place-name and the emigration story and invented a name for the leader of the emigrants.
There are legends linking Torsburgen to Thor, describing how he
could look out from its highest point over the sea, and of the god
avenging himself on the farmer who dared to try to build on it; cf.
Nihlén, 1975, 82–85.
K Traces of emigrants abroad
When the author describes the temporary settlement on Dagö
(Estonian Hiiumaa), he mentions a fortification, which ‘enn synis’.
It is possible that the author himself visited Dagö, but if not, his
knowledge of the fortification must either have come from an oral
tradition or from a written account. According to information from
the State Historical Museum in Tallinn (private communication),
however, no such construction is now evident. The fortress, although it might still have been extant at the time of writing, and
may have attracted oral tradition, seems not to have survived.
If it is accepted that there was a forced emigration from Gotland
in the fifth century, or at some other time, it is perhaps natural that
it would have been eastwards, and there is no doubt that parts of
Estonia have been settled by Swedish speakers at various times.
The two large islands off the coast of Estonia, Dagö and Ösel
(Estonian Saaremaa), lie north-east of Fårö, in the mouth of the
Gulf of Riga, and this may have been a more likely direction to take
than to the nearer coast of Latvia. One might also speculate upon
the reason that the emigrants could not stay there. Apparently their
numbers must have been so great that the area where they landed
was not able to support them and some, but not all, continued
eastwards. The island of Dagö is not large, smaller in area than
Gotland, and much of the centre is low-lying. It is possible to
imagine that it would not have supported a large influx of people.
The author’s sketch of the onward journey to Greece follows the
route customary for the time; see Note to 4/6.
L The tricking of the king
This passage distinguishes the emigration episode as told in Guta
saga from the more generalised accounts in the written sources
discussed above, pp. xxv–xxvii, and contains such a remarkable
number of alliterative phrases (so fierri foru þair, baddus þair
byggias, ny ok niþar, maira þan ann manaþr, þissun þaira viþratta

and so on) that it seems probable that some lost poetry lies behind
the story. If so, it is likely to have been of the orally-transmitted
variety. One would expect to find parallels to the episode of the
word-play used to trick the Byzantine emperor in ballads or folktales, the purpose of them being to show the superiority of one
group of people over another. The emigrants from Gotland are in
this case seen to outwit the monarch in one of the centres of learning
of the then known world. The fact that the empress is involved in
the dispute and successfully intercedes on behalf of the immigrants
perhaps reflects a Scandinavian social pattern, in which women
were more the equals of men than in other parts of Europe. Examples of influential women may be found in Landnámabók and in
several of the Icelandic sagas, for example Laxdœla saga. No close
parallels to this story have come to light, but there are similar tales
extant of ordinary people tricking monarchs (e. g. the ballad of
King John and the Abbot of Canterbury (Child no. 45 B) where the
disguised shepherd says, in reply to the king’s ‘Tell me truly what
I do think’, ‘You think I’m the Abbot of Canterbury’). There are
also a number of land-claiming tricks, for example ones in which
permission to claim only as much land as could be covered by a
hide is circumvented by cutting the hide into a thin strip and using
that to encircle the land claimed. One version of the story relates
that Birger Magnusson, who had been beaten by the Gotlandic
farmers at Röcklingebacke in 1313, was taken to Visby. There he
asked for as large an amount of land as a calf-skin would cover.
When permission was granted, he had the calf-skin cut into strips
and with these surrounded a considerable area, on which he had an
impressive royal residence built. This is said to be the origin of
Kalvskinnshuset in Visby, but this is probably more likely to have
been built as a symbol of Swedish power by Magnus Ladulås, who
was in a much stronger position than Birger. No king has ever lived
there and there are several more likely explanations for the name;
cf. Pernler, 1982. The motif itself probably goes back to ancient
methods of measuring land, perhaps with a ceremonial aspect; cf.
Söderberg, 1959, 48–49. Another example of the trick is related by
Saxo, in relation to Ívarr, the son of Ragnar loðbrók, and King Ella
of the Danelaw. Ívarr cut a horse-hide or ox-hide into narrow strips
and so was able to claim the land on which London was founded;
cf. Saxo Grammaticus, 1931–1957, I, 263; Ragnars saga loðbrókar,
1954, ch. 16–17.

A further clue to the origin of the particular tale in Guta saga
might lie in its final sentence: the people there still have ‘some of
our language’. To which people does this statement refer? Not the
Greeks, certainly, but perhaps the Goths or Getae, and this provides
another perspective on the emigration story. Weibull (1963, 33),
suggests that the reason for the choice of destination by the author
of Guta saga is that the Getae, who lived on the borders of the Byzantine empire, Grekland in Scandinavian sources, were linked to the
Goths and thus, by implication, to Gotland. It was, in other words,
an attempt at a folk-etymology of the name of Gotland. According
to one of what Gust Johansson (1968, 4) calls the ‘norröna trosartiklarna’, the Swedes of the Viking Age were directly descended
from those peoples who moved into the area as the ice drew back.
Johansson challenges this and favours the idea of the later invasion
by the Goths via Finland. This would appear to turn the whole of
the Gotlandic emigration story on its head, as it is at about the time
of the supposed exodus that Johansson assumes that the Goths,
beaten westwards by the Huns, moved into Sweden. The claim that
some people of eastern Europe ‘still have some of our language’
would then refer to the source of the language in Gotland. It has
been remarked that there are similarities between Gutnish and
Gothic; cf. Bugge, 1907. It is unlikely, however, that the author
himself would have been able to make the comparison, so consideration must be given to what justification the author can have for
the statement. It could be merely an invention to complete the
narrative, or some report might have come back with later travellers
to Constantinople or Jerusalem of a people they had met in the east,
who spoke a language reminiscent of their own, namely Gothic.
This in turn could have led to the invention of the emigration story,
based on the tradition current amongst the Black Sea Goths of their
origins on the island of Scandza, as recorded by Jordanes (1997,
33, 37, 81); cf. Tacitus, 1914, 195; GU, x–xi; Wolfram, 1988, 36
and references. Alternatively, the incident could merely be an
adaptation of a later movement eastwards with a resultant integration
of culture and language. Wessén, however, thinks it more likely
that there already existed an emigration tradition, and that this was
combined with a tradition amongst the Black Sea Goths, relating to
their origins, either by the author of Guta saga or earlier. Until the
end of the eighteenth century, there were the remnants of an East
Gothic community on the Crimean peninsula; see SL IV, 300.

M Heathen beliefs and practices
The heathen beliefs and practices described parallel to a degree
those proscribed in Guta lag (GLGS, 7). The source of the author’s
information could have been tradition, but there are a number of
written accounts which, while they would not have been the specific
ones used by the author, might suggest that his information came
from written material. Belief in sacred groves (hult) is recorded by
Tacitus (1914, 51, 190) and in Adam of Bremen’s description of
the temple at Uppsala (1961, 471–477) amongst others, and is so
well documented that no special source need be sought for this
piece of information. A respect for the howes (haugar) of ancestors
is also a commonplace and the numerous stories in Norse literature
of magical events associated with burial mounds are ample evidence of a cult related to them.

One of the most dramatic of the Gotlandic picture-stones (Hammars
I, preserved in the Bunge museum near Fårösund in the north of
Gotland, but originally from Hammars in Lärbro parish in northeast Gotland) shows what appear to be preparations for a human
sacrifice, with a figure lying across what seems to be an altar.
Gustaf Trotzig (GV, 370–371) writes of the figure, who is apparently being threatened by a spear, that he is particularly badly
placed (‘ligger illa till’). It might be significant that the potential
victim is considerably smaller than the other figures depicted; see
Lindqvist, 1941–1942, I, fig. 81; II, 86–87. Beyond him an armed
man seems about to be hanged, once the branch to which he is tied
is released, although Trotzig asks why, in that case, he is armed.
Adam of Bremen gives an account of human as well as animal
sacrifice at Uppsala, so one could accept that this picture-stone
carries evidence of heathen practice as well as belief. On a stone
from Bote, in Garde parish in central Gotland, there is a procession
of men who appear to have ropes around their necks and could be
about to be sacrificed. This particular scene is, however, open to
other interpretations. According to Guta saga human sacrifices
were offered for the whole of the island, whereas the thirds had
lesser sacrifices with animals, and there were also sacrifices on a
smaller scale more locally, possibly centred round the home of an
influential farmer, which later became the centre of a parish; cf.
Schück, 1945, 182. This structured organisation could well be a
later imposition of order upon what was a much more haphazard

arrangement, but there is no evidence for this either way. Steffen
(1945, 232–239) argues that the treding was a medieval division
and not a prehistoric one, with the original division of the island
being into two, but Schück (1945, 179–180) disagrees. Hyenstrand’s
study of the subject is referred to above, p. xxv.
Ibn Fadlan describes in detail the sacrifice of a servant girl at the
cremation of her Rus master, and the Russian Primary Chronicle
has a reference specifically to the sacrifice of their sons and daughters
by the people of Kiev, to idols set up by Vladimir, Jaroslav’s father;
see Birkeland, 1954, 17–24; RPC, 93–94. The victims of human
sacrifice were often slaves, criminals or prisoners of war, and the
means of death was frequently hanging as described by Ibn Rustah
and Adam of Bremen; see Birkeland, 1954, 16–17; Adam of Bremen,
1961, 471–473. The king, who represented a god, was sometimes
sacrificed in time of particular hardship, for example if the harvest
failed, and Håkon jarl offered his son Erling during the Battle of
Hjørungavåg in 986.
The subject of the extent and significance of human sacrifice is
discussed by, amongst others, Mogk and Ström. Mogk (1909, 643)
summarises his opinion as that the Germanic sacrifice was not an
act of punishment and that a cult act was involved, whereas Ström
(1942, 277–278) does not regard the death penalty as sacred, but
thinks that superstitions related to the act of killing led to rituals,
which gave a quasi-religious appearance to the deaths, making
them appear to have been self-inflicted. This does not, however,
explain the sacrifice of the king by the Swedes in time of need; cf.
Gordon, 1962, 256–257. Whichever is the case, the author of Guta
saga must have been aware of heathen traditions, since laws forbidding them were incorporated into Guta lag.

N Gotland’s treaty with Sweden
The successful negotiations by Avair Strabain with the king of
Sweden and the resulting treaty do not appear in any other known
source, but in King Alfred’s ninth-century translation of Orosius’
History of the World there is a description by a traveller named
Wulfstan of a voyage across the Baltic from Hedeby to Truso.
Wulfstan records that Gotland belongs (hyrað) to Sweden; see
Orosius, 16, line 28. This must be treated with caution, since he
also states that Blekinge did (Orosius, 16, line 27), and certainly by
the eleventh century the latter belonged to Denmark. On the other

hand, Rydberg (STFM I, 40) uses Snorri’s narrative in Heimskringla
to argue that Gotland was independent of Sweden at the time of
Olaf Skötkonung and Olaf Tryggvason, that is in the tenth century.
Gotland was said by Snorri in Óláfs saga Tryggvasonar (ÍF XXVI,
254–255) to have been the subject of a Norwegian attack (unlikely
if Gotland belonged to Sweden, as the two kings were allies) and
Rydberg’s argument, based on this account and another in the same
saga (ÍF XXVI, 337), places the treaty after the time of Olaf
Tryggvason. From the dating of a runic inscription on the Torsätra
stone in Uppland (U 614), it appears that some sort of tribute was
being paid to Sweden in the second half of the eleventh century.
The inscription records that Skuli and Folki had the stone raised in
memory of their brother Husbiorn, who fell sick abroad (usiok uti
‘vas siukR uti’) when they were taking tribute in Gotland. It is dated
to the 1060s or 1070s on account of its attribution to the runemaster Vitsäte, who appears to have been active about this time; see
Jansson, 1987, 88. Codex Laur. Ashburnham, the so-called Florensdokument, dated to circa 1120, mentions Guthlandia as one of the
‘insulae’ (literally ‘islands’, though the list includes non-insular
districts) of Sweden, but as it contains a number of very obvious
errors, its testimony on this point must be questionable; cf. Delisle,
1886, 75; DS, Appendix 1, 3, no. 4; Tunberg, 1913, 28; GV, 449–451.1
Whenever the treaty was negotiated, it seems possible that the
author of Guta saga had access to some written information about
the arrangements as they stood in his time and that there was then
some annual tax being paid to the Swedish crown. In 1285, King
Magnus Birgersson Ladulås issued an edict that each year the
Gotlanders should pay a levy tax in addition to the tribute, whether
or not a muster of ships were commanded. This seems not to have
been the case as described in Guta saga, where a levy tax is only
1 Various alternative theories have been advanced about the dating of the
incorporation of Gotland into Sweden. Nerman (1923, 67; 1932, 163–167;
1963, 25) argues from archaeological evidence of periods of disturbance
on Gotland and from finds at Grobin, Latvia, where Swedish and Gotlandic
artefacts from similar periods were found side by side, for an early dating,
around 550. Wessén, however, rejects this in favour of a date not long
before St Olaf’s visit in 1029; cf. SL IV, 306. Lindqvist (1932, 78) suggests
that the ninth century is a more likely period for the incorporation to have
occurred, since this was a time of Swedish expansion, and would have
offered advantages to Gotland of trade with the East.

demanded if the Gotlanders for some reason fail to provide the
ships for the levy, so the annual tax would have been separate.
These taxes would have been in silver, a material not available in
the migration period; see Nerman, 1932, 167. If, as has been
suggested by Sjöholm (1976, 108), Guta saga was written as a
legendary history with the purpose of arguing the case for Gotland’s
autonomy, it would be necessary to demonstrate that the agreement
had been first entered into freely and not under duress, as a symbiotic relationship that did not involve Gotland relinquishing its
sovereignty. Written sources for an early agreement seem unlikely
and the possibility of oral sources is further discussed below. The
treaty terms themselves probably relate more closely to those of
the author’s day than to those of 200 years or more previously and
the change from the preterite to the present tense in the text might
support this argument.
If the interpretation put by Wessén on the expression fielkunnugr
is to be accepted, and Avair Strabain was skilled in ‘magical’ arts,
this would, according to Wessén, place the treaty in the heathen
period; see SL IV, 306 and Note to 6/9–11. Any details about it
must therefore have come from oral rather than written tradition,
possibly a narrative verse. There is no hint as to where the author
found his story, and no other record of an Avair, but it is possible,
one might suggest, that he had heard poetry concerning a muchrespected heathen who might have acted as intermediary in such a
negotiation. Schütte (1907, 83) compares Avair Strabain in Guta
saga to a character in a tale told about Charlemagne who succeeded in getting a law agreed upon where others had failed. He
suggests that this could have been a model for the episode in Guta
saga. The man in this story later disappears without trace, as
mysteriously as he appeared, but the information about Avair seems
to be more circumstantial. He is given a home parish, a wife and
a son, and extracts a promise of compensation should his mission
miscarry. As Wessén remarks, the alliterative phrase faigastan ok
fallastan (‘doomed and ill-fated’) suggests an oral tale behind the
speech Avair delivers, if not behind the whole story; see SL IV,
306. The use of such phrases is not, of course, confined to poetry
and there would be no reason for particularly assuming that a lost
verse lay behind this episode, were it not for the proliferation of
alliterative phrases throughout the passage, for example: siþan
sentu gutar sendimen, fikk friþ gart, gierþi fyrsti friþ. The use of

parallelism is a further indication of possible poetic origin for this
part of the narrative. Similarly, the details of the actual treaty
contain evidence of an oral source. Phrases such as frir ok frelsir,
hegnan ok hielp and steþi til sykia may well have had their origin
in verse, although they may have been in written form when used
by the author of Guta saga, possibly as a set of legal formulae.
It is dangerous to rely on place-names to support oral traditions
and there seems little doubt that names such as Ava, Avagrunn and
Avanäs lack any connection with Avair, particularly as they are all
on Fårö, and as the names are repeated on the Swedish mainland.
A village named Awirstadha in the parish of Askeby, Östergötland,
was mentioned in a manuscript from 1376 (designated 9/10 in
Linköpings stifts- och landsbibliotek, but probably destroyed in a
recent fire); see Sveriges medeltida personnamn, 1974– , s. v. Aver.
Hilfeling (1994–1995, I, 184) records the existence of a Strabeins
grav in Alva parish and provides a sketch of it, but the designation
of this kämpargrav to Avair is, as he says, not historical. Even if it
is accepted that Avair was a historical figure, which is by no means
certain, firm evidence is still lacking for the dating of the treaty.
One also wonders if, in the figure of Ívarr beinlauss of West Norse
tradition, there is any sort of parallel to Avair Strabain.

Preserving Our Past, Recording Our Present, Informing Our Future

Ancient and Honorable Clan Carruthers Int Society CCIS  LLc

carruthersclan1@gmail.com              carrothersclan@gmail.com

You can find us on facebook at :

https://www.facebook.com/carrutherscarrothers.pat.9

https://www.facebook.com/CarruthersClan/

https://www.facebook.com/CarruthersClanLLC

Disclaimer Ancient and Honorable Carruthers Clan International Society CCI