Chapter 5
Interdisciplinary Modelling of Viking Age
and Medieval Settlement in the Mosfell Valley
S
cholars have long looked to rich and vivid medi-
eval Icelandic narrative texts to understand the
initial settlement of Iceland, which began around
ad  870. The Book of Settlements, Landnámabók, re-
counts the stories of over four hundred settlers who
founded and named farmsteads, many of which are still
known by the same name today. In some cases, the text
only identifies the heads of households, but in others, it
mentions attached farm workers or lists followers of the
primary settlers as they establish subsidiary farmsteads.1
Íslendingabók (The Book of Icelanders) contains a shorter
and in many ways complementary account of the settle-
ment of Iceland. Íslendingabók states, for instance, that
the island was albyggt (completely settled) by ad  930,
sixty years after the arrival of the first Norse migrants.2
Scholarly estimates of the total population inhabiting
the island by ad 930 vary, but most range from 10,000 to
20,000 people.3 The accounts preserved in Íslendingabók
and Landnámabók texts are widely believed to be based
on oral tradition and to contain significant historical
value, although not everyone agrees to what extent.4
1 Over six hundred individual settlement period farmsteads are
named. See Gunnar Karlsson, The History of Iceland, p. 14.
2 Íslendingabók, Landnámabók, ed. by Jakob Benediktsson,
p. 9. The passage states: ‘Svá hafa spakir menn sagt, at á sex tegum
vetra yrði Ísland albyggt, svá at eigi væri meirr síðan’ (‘Wise men
have said that after sixty years Iceland was completely settled, such
that no more settlers came after that’).
3 Gunnar Karlsson says no fewer than 10,000 settlers; see
Gunnar Karlsson, The History of Iceland, p. 15; Byock, Viking Age
Iceland, p. 9.
4 For an argument that Landnámabók is not to be trusted
Davide Zori <dmzori@gmail.com> is Assistant Professor at Baylor University in the Baylor Interdisciplinary Core. He is Field Director of the Mosfell
Archaeological Project and has been Staff Researcher at the Center for Medieval and Renaissance Studies, University of California, Los Angeles.
Viking Archaeology in Iceland: Mosfell Archaeological Project, ed. by Davide Zori and Jesse Byock, CURSOR 20 (Turnhout: Brepols, 2014)
10.1484/M.CURSOR-EB.1.102212
Davide Zori
Maps based on Landnámabók are frequently used to
show the distribution of initial settlement in Iceland and
to demonstrate a range of early settlement patterns, such
as the concentration of habitation along the coast and
particularly dense settlement clusters in the south, the
southwest, and the northern fjords.5
Significant questions remain, however, about the
comprehensiveness of the settlement list in Landnáma-
bók, as well as the rate and density of Iceland’s early
settlement and the statement in Íslendingabók that the
land was ‘albyggt’ (‘completely settled’) by 960.6 Albyggt
is usually interpreted to mean that all land had been
claimed. However, the density of farmsteads on the land-
scape was likely to have been quite low at this early time
in Iceland’s colonization. The Icelandic Family Sagas
(Íslendingasögur) can help to provide a model of the set-
tlement and the partitioning of the land that occurred
subsequently. In the earliest years of settlement, whole
fjords and regions were claimed by single settlers, or
landnámsmen, such as Ingólfr Árnarson, Helgi Magri
(the Lean), and Auður Djúpúðga (the Deep-minded).
The original claimants then gave large parcels of land to
their supporters. The area of Helgi Magri’s initial land
claim would be parcelled out to 20–30 different set-
tlers, and by the eighteenth century, there were approxi-
mately four hundred and fifty farms within Helgi’s origi-
for true information regarding the settlement period, see Adolf
Friðriksson and Orri Vésteinsson, ‘Creating a Past’, pp. 139–61.
5 See examples of maps in Byock, Viking Age Iceland, p. 85. See
also Árni Daníel Júlíusson, Ólafur Ísberg, and Helgi Skúli Kjartans-
son, eds, Íslenskur sögu atlas 1.
6 Íslendingabók, Landnámabók, ed. by Jakob Benediktsson, p. 9.
pp. 55–79
BREPOLS PUBLISHERS
56
nal settlement area.7 A comparable case is Skallagrímr
Kveldúlfsson’s land claim in Borgarfjörður, recorded in
Egils saga. The saga describes in detail how Skallagrímr
established allies and supporters to take advantage of
the varied natural resources within his original land
claim. This account, which can be called the ‘Skallagrímr
model’, is a template for how emergent chiefs might have
used bestowal of land to consolidate their power in the
newly settled Icelandic landscape.8
Interdisciplinary studies, drawing in archaeological
data and place names, have the potential to provide a
more nuanced understanding of the settlement of indi-
vidual regions of Iceland. Such studies can be used to test
whether the texts accurately document the density and
number of Settlement Period (ad 870–930) farmsteads
in a region. By providing an indication of settlement
hierarchy and status differentials between contempora-
neous farmsteads, interdisciplinary studies furthermore
have the possibility of testing whether the Skallagrímr
model can be applied to understand the early settlement
patterns of other regions in Iceland.
In this chapter, I provide an interdisciplinary case
study of the settlement of Iceland by examining the evi-
dence available for the Mosfell Valley, including textual
data, place names, archaeological survey, sub-surface
testing, and extensive excavations conducted at the
site of Hrísbrú by the Mosfell Archaeological Project
(MAP). For this valley, Landnámabók identifies only
one site, Skeggjastaðir, as having been established in the
Settlement Period, while the Icelandic saga corpus iden-
tifies one additional site, Mosfell, as a chieftain’s farm-
stead before ad 1000. I begin with an examination of
texts from the pre-modern period to provide clues about
the patterning and hierarchy of the valley’s settlement.
Principles of Norse place-naming practice are then used
to suggest a model for the sequence of settlement in the
Mosfell Valley. The place name evidence suggests that
the medieval texts are rather incomplete in their record
of the farms in this particular valley. The place names
suggest that in addition to Mosfell, Helgadalur and
Leirvogstunga were Settlement Period farms, while the
-staðir farms (i.e. Skeggjastaðir, Hraðastaðir, Æsustaðir)
date to the first few generations after the initial set-
tlement. Finally, archaeological data collected by the
Mosfell Archaeological Project are used to test the place
name model, demonstrating that the valley was much
Gunnar Karlsson, The History of Iceland, p. 15.
7
See, for example, Byock, Viking Age Iceland; Orri Vésteinsson,
8 that in Iceland starts with the human settlement (c. AD 870) and lasts
McGovern, and Keller, ‘Enduring Impacts’.
Davide Zori
more quickly and densely settled than suggested by the
textual records.
Historical texts, maps, and aerial photographs pro-
vide a rich background to the archaeological work in
the Mosfell Valley. Historical land registers, including
a comprehensive one from ad 1704, identify the tradi-
tional farms in the valley, while medieval church charters
indicate the existence of some of these farms at earlier
dates. Historical maps from the late nineteenth and early
twentieth century provide a visual representation of the
pre-modern farms that correspond to the farms identi-
fied in the land registers. In 1916, all farmsteads in the
valley were drawn in plan, fortuitously supplying a view
of the old farmsteads as they appeared when constructed
of turf and stone, before the construction of concrete
structures and prior to the effects of mechanized farm
equipment on the landscape. Historical aerial photo-
graphs from every decade since 1950 help us understand
local landscape change and particularly the effects of
the recent surge in construction activity in the valley.
Nineteenth century accounts of local ruins, archaeo-
logical remains, and oral traditions in the Mosfell Valley,
many of which use the contemporary built landscape as
a starting point of their descriptions, provide vital clues
to cultural remains that are no longer visible today. Some
of the features described in the pre-modern sources are
depicted in the maps and early aerial photographs, mak-
ing these three datasets even more informative when
used in combination with one another.
Because of the significant deposition of eroded soils,
Icelandic archaeology in lowland and valley environ-
ments such as the Mosfell Valley requires subsurface
investigations. No surface scatters of finds exist and
structural ruins from the Medieval Period (settlement
to ad 1500)9 are very rarely visible on the surface. As
a result, successful identification of medieval remains
is difficult to achieve and comprehensive identification
of all sites in a region even more so. The search for early
Icelandic farm sites, however, is facilitated by the fact
that the locations of the early farms and main habitation
structures have been very stable through time. From the
start, we designed the MAP research strategy with the
9 ‘Medieval’ is sometimes in Iceland and Scandinavia used to
refer narrowly to the period following the Viking Age. The term is
employed here and thoughout this book to refer generally to the
wider Middle Ages, stretching from the end of the Roman Empire
until the Renaissance or Reformation, whichever comes first. The
Viking Age is here considered as a sub-period of the Middle Ages
until c. ad 1100.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
anticipation that the Viking Age and medieval farms
would be close to or even under modern farm buildings.
The results of a survey project in Skagafjörður employing
sub-surface coring and geophysics provides support for
our assumption by showing overall stability of farmstead
location in that survey area.10
The archaeological work of MAP has consisted of
open area excavation, test trenching, subsurface coring,
and surface survey and mapping. Aiding the temporal
control of this work are a series of volcanic ash and ejecta
layers, or tephras, from eruptions of known dates that
are found in the stratigraphy in the Mosfell Valley.11 The
tephra layers are distinguishable from one another by
colour, thickness, grain size, and stratigraphical location.
The tephra layers in the valley and the dates at which they
were deposited have been identified and are as follows:12
– Landnám Tephra – ad 871 ± 2
– Katla-R Tephra – c. ad 920
– Eldgjá-1 Tephra – c. ad 934
– Medieval Tephra Layer – ad 1226
– Katla1500 – c. ad 1500
– Katla 1721 – c. ad 1721
Volcanic tephra layers often appear in the turfs used for
the construction of medieval Icelandic houses. These
tephra layers in turfs therefore provide a terminus post
quem for the construction of the turf structures. Soil
thickening in most areas of the Mosfell Valley from the
late ninth century until the mid-thirteenth century ad
has been approximately 0.3–0.6 mm per year, although
up to 2 mm per year has been observed. Under the thick-
ening rate of 0.4 mm/year, 1 cm of soil would take about
twenty-five years to form.13 Under these conditions, the
period during which a tephra layer might have been incor-
porated into the harvested turfs is probably in the range
of one hundred years. Turf is usually cut from the same
location, meaning that once turf containing a particular
10 Bolender, Steinberg, and Damiata, ‘Farmstead Reorganiz-
ation at the End of the Viking Age’.
11 For a general discussion of tephra layers and their use for
archaeological research in Icleand, see Byock and Zori, Chapter 1 in
this volume.
12 Magnús Sigurgeirson, ‘Appendix C: Archaeological Investi-
gation at Hrísbrú in Mosfelldalur’, section 7 (pp. 45–46).
13 Magnús Sigurgeirson, ‘Appendix C: Archaeological Investi-
gation at Hrísbrú in Mosfelldalur’.
57
tephra is cut from a wetland, no other turfs subsequently
taken from that area will contain that same tephra.
With the precise dating provided by these tephra
layers, the archaeological remains in the valley reveal a
landscape that was much more intensively settled by the
earliest Norse migrants than indicated by the textual
evidence. This chapter demonstrates that archaeology
is better suited than the medieval textual sources for an
investigation of the establishment, spatial patterning,
and settlement histories of the farmsteads in the Mosfell
Valley. Nevertheless, this does not mean that texts and
place name data do not add valuable information to the
story, particularly in the Mosfell Valley, where the tex-
tual sources are abundant and the place names in the
landscape have remained stable through time.
The Textual Record of the Early Farms
in the Mosfell Valley
With the drainage of swampy areas of the Mosfell Valley
and the mechanization of agricultural equipment, the
modern period has seen more land come under culti-
vation and the establishment of new farms throughout
the valley. Overflow from the growing urbanization of
Reykjavík and the town of Mosfellsbær has also brought
non-farming residents into the valley. These new arriv-
als settled mostly on the wet and recently drained val-
ley floor. Land registers from the pre-modern period
identify the old farms pre-dating this twentieth-century
development and population growth and provide a use-
ful starting point for discussing the medieval farms.
Medieval narrative sources and late medieval land char-
ters are less exhaustive, but help to identify the age, his-
tory, and importance of the farms and their names.
Among the pre-modern land registers, the most com-
prehensive and useful for identifying the pre-modern
farms of the valley is Árni Magnússon and Páll Vídalín’s
Jarðabók from 1704.14 The Danish king commissioned
this register to take account of his Icelandic lands. Table
5.1 on the following page lists the farms in the Mosfell
Valley and their estimated contemporary value accord-
ing to the 1704 Jarðabók.
Ten of the twelve farms listed in Jarðabók still exist
today. The exceptions are Bakkakot and Hlaðgerðarkot,
which were small tenant farms owned by the Mosfell and
Norður Reykir farms respectively. The sources indicate
that Bakkakot was established in the seventeenth cen-
14 Jarðabók Árna Magnússonar og Páls Vídalíns.
58 Davide Zori

Table 5.1. The farms in the Mosfell Valley as recorded in the Jarðabók by the Viðey monastery, situated on an island in the bay
land register from 1704. The spelling of the modern farm west of the Mosfell Valley.17
names is provided in parenthesis if different from the 1704
form. The column on the right records the total worth of
Laxnes, Leirvogstunga, and Minna Mosfell first appear
each farm in the long-hundred system (i.e., one ‘hundred’ is in Fógetareikningar 1547–52. Fógetareikningar comprises
120). the king’s magistrate’s account of the lands owned by the
king after the Reformation, during which the Danish king
Farm Name Worth (in hundreds)
took control of lands previously owned by the Catholic
Backakot (Bakkakot) recorded with Mosfell Church, including the rich Viðey monastery.18
Helgadalur 10 These documents provide terminus ante quem dates
for the establishment of the farms in the Mosfell Valley,
Hladgierdar kot (Hlaðgerðarkot) 10
but the farms could of course be older. In fact, some
Hradastader (Hraðastaðir) 10 farms that are not recorded until 1704 may be older than
Hrijsbru (Hrísbrú) 28 farms that appear in the 1395 Viðey charter or the six-
Laxnes 15 teenth-century Fógatereikningar, since these latter two
are not comprehensive land registers and include only
Leirvogstúnga (Leirvogstunga) 28
the lands owned by the monastery and the Danish king,
Minna Mosfell 10 respectively. Closer analysis of the individual farm names
Mosfell 30 provides more information on the relative and absolute
Nordur Reykier (Norður Reykir) 10
dating of the Mosfell Valley farms, as well as the settle-
ment order and hierarchy among the farms.
Skieggiastader (Skeggjastaðir) 5
Æsutader (Æsustaðir) 10
A Settlement Model Based on Place Names
The place names of the Mosfell Valley recorded in medi-
tury and abandoned by the middle of the nineteenth eval sagas, church charters, tax registers, early modern
century. Hlaðgerðarkot is probably older. The use of maps, and the rich local oral tradition reveal a wealth of
the habitative generic -kot in farm names began rela- information about the economy, cultural history, and
tively late in Iceland and first appears in the sources in the evolution of power structures subsequent to the
the fourteenth century, after which it become popular in arrival of the Norse settlers in the late ninth century.
names of small attached tenant farms such as Bakkakot The place names of farms, roads, fields, mountains, val-
and Hlaðgerðarkot.15 leys, rivers, fjords, and other cultural and natural features
Five of the farms listed in the 1704 Jarðabók appear in the Icelandic landscape hold a rich potential for the
in medieval sources pre-dating ad  1400: Helgadalur, understanding of early Icelandic society. The value of
Hraðastaðir, Hrísbrú, Mosfell, and Skeggjastaðir. Only place names for studies of historical geography, settle-
Hrísbrú, Mosfell, and Skeggjastaðir, however, are men- ment patterns, and belief systems is widely recognized
tioned in sources concerning the Free-State Period and is based on the principle that the original meaning
(ad 930–1262). Landnámabók holds that Skeggjastaðir of a place name often sheds light on the inhabitants and
was the valley’s first farm, established by and named the function (economic, ritual, and/or political) of that
after the valley’s original settler or landnámsmaðr, site at the time of its naming.19
Þórðr Skeggi. In several sagas, including Egils saga, Icelandic place names are easily comprehensible
Gunnlaugs saga ormstungu, and Hallfreðar saga, and because they all derive from the Norse settlers who
in Landnámabók, the Mosfell farm is identified as the arrived as a unified cultural group with a common lan-
seat of the Mosfellsdælingar chieftains and the centre of guage in a relatively recent and single-period migration.
power in the region.16 Helgadalur and Hraðastaðir first
appear in a charter from ad 1395 of the lands owned
17 Diplomatarium islandicum, ed. by Jón Sigurðsson and

15Svavar Sigmundsson, ‘Place-Names in Iceland and Shetland’, others, iii, 598.

pp. 209–15. 18 Diplomatarium islandicum, ed. by Jón Sigurðsson and

16 For background on the Mosfellsdælingar chieftains, see others, xii.

Byock and Zori, Chapter 1; and Byock, Chapter 3, in this volume. 19 Bühnen, ‘Place Names as an Historical Source’, p. 65.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
Until modern times, Icelandic society remained primar-
ily a conservative, monolingual, and monoethnic soci-
ety, meaning that the place names have remained unu-
sually stable. A large corpus of early written documents
recording these names testifies to the preservation of the
names and settlement patterns over time. The stability
of Icelandic settlement patterns and the very recent and
limited scope of urbanization have left a social landscape
in which the vast majority of the farms mentioned in
Landnámabók still exist today in virtually the same place.
Scandinavian place name scholars have developed a
body of theory that allows for the use of place name types
in the reconstruction of settlement order and social rank-
ing. The most significant observation for the study of the
Mosfell Valley is that there is a corresponding pattern
between names and the order in which they were estab-
lished. In the Viking world, the first Scandinavian farms
in an area almost always take their name from large topo-
graphical features. Similarly, in the Norse expansion in
the north Atlantic, ‘[t]he primary sites are named after the
most prominent, mostly coastal, topographical features’.20
Secondary settlements frequently contain habita-
tive generic suffixes.21 One of the most common generic
place name elements of the Viking Age that appears
20 Kruse, ‘Norse Topographical Settlement Names on the
Western Littoral of Scotland’, p. 105.
21 Place names can be broadly divided into two types:
topographical names and habitative names. Topographical names
(or nature names) describe a topographical feature, almost always
natural, but also possibly cultural. Habitative names (or culture
names) denote inhabited places, such as farms, villages, enclosures.
59
Map 5.1.
The primary settlements in
the Mosfell Valley. Analysis
of the place names suggests
that Mosfell was the first
farm established in the
valley, followed soon after by
Helgadalur, and Leirvogstunga
(Tunga). The original Mosfell
farm, as depicted on the map,
was located at the site of the
current Hrísbrú farm
(Davide Zori).
across the Norse North Atlantic colonies and frequently
in the Mosfell region is -staðir. 22 In Norway, schol-
ars have established that -staðir names with personal
appellatives as the first element date to the Viking Age,
whereas -staðir names with a first element referring to
natural topography are more likely to date to the pre-
Viking period.23 The -staðir place name largely went out
of style in the eleventh century,24 suggesting that -staðir
names in Iceland should be of Viking Age origin. In
Iceland, the generic -staðir appear to have been used for
secondary farms established in areas originally inside the
borders of the primary settlement farms.25 In such cases,
the personal name used in these constructions could be
either the name of the owner of the land or the name of
the occupant/tenant of the farm. The primary settlers of
Iceland probably gifted or rented subsidiary -staðir farms
to followers, dependents, or new-comers, indebting these
inhabitants economically and politically to the origi-
nal settlers.26 In fact, the term -staðir originally meant
‘land parts’ and the usage of the plural form staðir rather
than the singular staðr may be due to the fact that there
were several divided parts from each original farm.27
22 Kruse, ‘Norse Topographical Settlement Names on the
Western Littoral of Scotland’, p. 104; see also Gammeltoft, The Place-
Name Element Bólstaðr in the North Atlantic Area.
23 Andersen, ‘The Norwegian Background’, p. 17.
24 Olsen, Hvad Våre Stedsnavn Lære Oss, pp. 36–39.
25 Svavar Sigmundsson, ‘Íslensku staða-nöfnin’, pp. 238–48.
26 Svavar Sigmundsson, ‘Íslensku staða-nöfnin’, p. 240.
27 Svavar Sigmundsson, ‘Íslensku staða-nöfnin’, pp. 241–43.
60
The -staðir farms therefore provide evidence of early
divisions of large land claims.
Finally, tertiary farms have less prestigious names
derived from other settlement names or taken from
smaller natural or man-made features in the landscape.
As a group, these farms are much more difficult to
ascribe any clear date range. In the most general sense,
they can be tentatively dated sequentially as established
after the secondary settlements, and continue to be used
up to the present day.
I suggest that we can derive the following predictive
model of the regional sequence of settlement on the
basis of these place name studies:
1) Primary settlements established during the Norse
Viking Age migrations are named after prominent
landscape features. These farms should date from
the beginning of the settlement of Iceland around
ad 870 until approximately ad 930, at which point
Islendingabók declares that the island is fully settled.
2) Secondary settlements frequently contain habitative
suffixes, particularly -staðir, with a preceding per-
sonal name. These farms could have been founded
soon after the establishment of the primary farms
and up to the early eleventh century, at which point
the -staðir names cease to be used for new farms.
3) Tertiary settlements were established subsequent to
primary and secondary farmsteads and usually bear
names derived from other settlements or taken from
minor natural features or man-made structures al-
ready present in the landscape.
Davide Zori
Map 5.2.
The primary and secondary
settlements in the Mosfell
Valley. Place name analysis
suggests that the -staðir farms,
Skeggjastaðir, Hraðastaðir,
and Æsustaðir, were settled
before ad 1000 as farms
splintering from the lands
of the Mosfell farm
(Davide Zori).
Following this predictive framework, the place names
in the Mosfell Valley can be employed to propose a three-
stage model for the settlement order and hierarchy of the
farms in the valley. According to this model, the primary
farms in the valley include the three farms with macro-
topographical names: Mosfell, Tunga (Leirvogstunga),
and Helgadalur (Map 5.1). Before the eleventh century,
three secondary settlements, Skeggjastaðir, Hraðastaðir,
and Æsustaðir, which employ the habitative suffix -staðir
preceded by a personal name, appear evenly spaced across
the landscape (Map 5.2). The tertiary farms, including
Hrísbrú, Laxnes, Minna-Mosfell, Norður Reykir, and
Hlaðgerðarkot, appeared in the late Middle Ages and
the early modern period (Map 5.3). The subsequent sec-
tions provide a closer analysis of each individual farm,
according to the place names, the textual evidence, and
the archaeological data.
The Primary Farm:
Apex of the Site Hierarchy in the Valley
The place name settlement model predicts that the farm
named Mosfell was the first settlement in the Mosfell
Valley. Mosfell, meaning ‘Moss-mountain’, is the name
of the valley’s most prominent mountain, located just
inland from the Leiruvogur and Kollafjörður bays.28 The
28 The first element (Mos-) refers to the green vegetation that
would have covered the mountain when the Viking Age settlers first
arrived. See Kålund, Bidrag til en Historisk Topografisk Beskrivelse af
Island, p. 47.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
farm of the same name lies at the foot of the mountain’s
southern slope. Strong supporting evidence that Mosfell
is the primary farm in the region is provided by the
inclusion of Mosfell as the appellative base in almost all
regional place names, including Mosfellsdalur (Mosfell
Valley), Mosfellssveit (Mosfell region), Mosfellsheiði
(Mosfell heath), and the modern Mosfellsbær (Mosfell
town). The mountain is a distinct and clear landmark in
the landscape that is prominently visible from the sea,
where it stands out between two valleys.
The question from the perspective of the landscape
and the archaeological remains is where exactly the origi-
nal Mosfell farm was located. The simple assumption
based on the stability of the Icelandic farms would be
that the original farm lies in the same place as the cur-
rent farm called Mosfell. The traditional theory holds
that the Mosfell farm originally utilized the entirety of
the southern slopes of the Mosfell Mountain, but later
in the Medieval Period this productive land was divided
for more intensive use between three farms: Mosfell,
Hrísbrú, and Minna-Mosfell.29
The available textual sources, however, indicate that
the original Mosfell farm moved from the current site
of the farm Hrísbrú to its present position. 30 Since the
29 Magnús Grímsson, ‘Athugasemdir við Egils sögu Skallagríms-
sonar’.
30 See Björn Thráinn Thórðarson, Chapter 2 and Jesse Byock,
Chapter 3 in this volume for more on movement of the Mosfell
farm. The relocation of farms and farm names was apparently not
un common and has been observed with other early farms such
as Hvammur in Norðurárdalur. In this case the old farm location
61
Map 5.3.
The primary, secondary, and
tertiary settlements in the
Mosfell Valley. The tertiary
sites that cluster on the
northern and southern slopes
of the valley were products of
the gradual intensification of
the agricultural production in
valley. In the twelfth century,
as depicted on this map,
the Mosfell farm changed
locations and the old farm
site was renamed Hrísbrú
(Davide Zori).
name of the farm and the identity of the farm’s inhab-
itants were inexorably linked, when the primary farm
moved, the place name Mosfell moved with it, and the
name Hrísbrú was given to the location of the old farm.
Egils saga provides a clue to the date of the relocation of
the Mosfell farm by stating that the church was moved
when the priest Skapti Þórarinsson lived at Mosfell. The
same Skapti is recorded in prestatal, a list of important
Icelandic priests from ad 1143.31
Hrísbrú: The Original Mosfell Site
The textual record suggests that the current farmstead
named Hrísbrú was the location of the original Mosfell
farm. Archaeological research corroborates this sugges-
tion with the excavation of a chiefly farmstead, including
a large longhouse, an early church and Christian grave-
yard, and a pre-Christian cremation site (Figure 5.1).
The combined evidence from place names, the sagas, and
archaeological excavations agree that the farm located
at Hrísbrú occupied the top of the settlement hierarchy
as the most successful, and in all likelihood, the earliest
farm in the valley.
The house excavated at Hrísbrú is a Viking Age long-
house built between ad 871 and ad 920/934, partially
rebuilt after 920/934, and abandoned in the eleventh
was usefully renamed Fornihvammur (Old Hvammur), see Jón
Jóhannesson, A History of the Old Icelandic Commonwealth, p. 33.
31 Byock and others, ‘A Viking-Age Valley in Iceland’,
pp. 195–218.
62 Davide Zori

Figure 5.1.
Aerial photograph from 2006 of
the Hrísbrú farmstead, showing
the locations of the Viking Age
longhouse, the conversion-period
church, and the cremation site
(Hulduhóll). North of the large
corrugated-steel Quonset hut and
east of the church excavations
is a deep sequence of cultural
layers that include the bulldozed
remains of the nineteenth-
century turf farm shown in the
drawing from 1916 (Figure 5.2).
The post-longhouse medieval
farmstead may lie at the bottom
of these cultural layers (Mosfell
Archaeological Project).

century (see Plates 1 and 3 and Figures 1.2, 1.3, and

1.5 for maps and photographs of excavated Hrísbrú
farmstead).32 The longhouse appears to be part of the
initial habitation of the Hrísbrú farm. Radiocarbon
dating of barley seeds collected from the house floor all
have a date range with a centre predating ad 1000, while
samples from the overlying midden deposits date from
the late tenth to early eleventh century (see Graph 1.1
and Figure 1.7). Skeletal remains from the churchyard
yielded dates from the late tenth to early eleventh centu-
ries. Beads recovered from the longhouse are all of types
common in the tenth to early eleventh centuries.33
Minor restructuring of the house took place after
ad 920/934, but the longhouse retained the same basic
form throughout its occupation. After the abandonment
of the house, no subsequent structures were built directly
atop the site, but it is clear from the midden deposits
Figure 5.2. Map of the Hrísbrú farmstead drawn in 1916 showing
the location and layout of the farm when it still consisted of the
traditional turf buildings that date back at least to the nineteenth 32 Landnám Tephra (ad 871 ± 2) in the all walls of the house
century. Two roads access the farm, one from the south and one from
indicates that it was built after ad 871. A thin black tephra layer,
the west that continues east towards the Mosfell farm. These old farm
which is either the Katla-R Tephra from 920 or the Eldgjá Tephra
buildings were partially bulldozed for the construction of the large
from 934, was found only in a section of repaired walls, indicating
hanger-like Quonset hut seen on the aerial photograph in Figure 5.1
that the original house was built before that tephra fell. Byock
(map by author based on 1916 original in Þjóðskálasafn Íslands).
and Zori, Farmstead Survey in the Mosfell Valley. For additional
discussion of dating of the house, see Zori, ‘From Viking Chiefdoms
to Medieval State in Iceland’.
33 See Chapter 10 by Elín Hreiðarsdóttir in this volume. The
typology and dating system used by Hreiðarsdóttir is derived from
Callmer, Trade Beads and Bead Trade in Scandinavia. See also Elín Ósk
Hreiðarsdóttir, ‘Appendix A: The Glass Beads from Hrísbrú 2008’.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
dumped into the abandoned and collapsed house that
domestic occupation of an unknown scale continued at
the Hrísbrú farm somewhere near the abandoned long-
house. It is likely that the post-longhouse farm shifted
to the location where the traditional farm was still posi-
tioned when it was drawn in 1916 (Figure 5.2). This
turf farm was partially bulldozed to make room for the
large repurposed hangar-like Quonset hut seen in the
aerial photograph from 2006 (Figure 5.1). Deep cultural
sequences in the mound behind the Quonset hut sup-
port this interpretation.
The sagas recount stories of chieftains living at
Hrísbrú/Old Mosfell around ad 1000.34 Independently,
the archaeological evidence supports the conclusion that
the Viking Age Hrísbrú farmstead was occupied by a
high status household. The length of the house, at almost
30 m measured externally, is among the longest exca-
vated in Iceland. The number of imported glass beads is
much greater than all other assemblages from Viking Age
houses in Iceland. The size of the house, combined with
the large number of barley seeds and high ratios of cattle
34 For the textual sources on the Hrísbrú/Mosfell chieftains,
see Chapter 3 by Byock in this volume.
63
Figure 5.3.
Map of the Mosfell
farmstead from 1916,
showing the traditional
turf farm, the desanctified
churchyard, and the old
road that leads west to
Hrísbrú and east to Minna-
Mosfell (map by author
based on 1916 original in
Þjóðskálasafn Íslands).
to caprine bones, suggest that the chieftains occupying
the house hosted patron-role type feasts for followers.35
The presence of an early Conversion Period (around
ad 1000) church at the farmstead is also consistent with
expectations of higher status households, which quickly
realized and utilized the potential ideological power of
Christianity.
The Current Mosfell Farm:
Medieval and Post-Medieval Evidence
Archaeological research at the current Mosfell farm has
demonstrated the presence of a medieval farmstead and
church that predate ad 1500, but no evidence has been
found of a Settlement Period farm at this location. The
presence and size of the medieval churchyard at Mosfell
strongly suggests that this farm was a power centre
after the farm had been moved here from Hrísbrú. The
archaeological remains of the medieval and pre-modern
35 Zori and others, ‘Feasting in Viking Age Iceland’, pp. 150–65.
See Chapter 12 by Zori and others for a presentation of the faunal
material from the Hrísbrú excavation.
64
farmstead at Mosfell have been greatly impacted over
the centuries as part of the frequent remodelling of the
landscape and the construction of multiple churches and
farm houses. As a result, a large portion of the medieval
layers appears to have been disturbed and/or removed.
Fortunately, an unusually detailed nineteenth-century
description of the archaeological evidence of the earlier
farms has been preserved in Magnús Grímsson’s account
of the Mosfell farmstead, written while he served there
as parish priest.36 The following analysis of the current
Mosfell farm draws heavily on old maps, historical refer-
ences, and aerial photographs to demonstrate the poten-
tial for such work when excavation is not possible.
The nineteenth-century farm buildings and church at
Mosfell can be seen as a functioning farm in the 1916
farm map and as ruins in the 1954 aerial photograph
(Figures 5.3 and 5.4). In 1965, this area was bulldozed
and the parson’s house was built in the same spot as the
earlier ruins (Figure 5.5). As seen in the aerial photo-
graph from 2006 (Figure 5.6), this area was again bull-
dozed and levelled to create the modern parking lot
east of the modern church. Grímsson writes that the
absence of other visible ruins in the nineteenth century
and the presence of large midden heaps outside of the
36 Magnús Grímsson, ‘Athugasemdir við Egils sögu Skallagríms-
sonar’. For more about Magnús Grímsson, see Chapter 2 by Björn
Thráinn Thórðarson in this volume.
Davide Zori
Figure 5.4.
Aerial photograph of the
abandoned Mosfell farm
in 1954. The old road,
which appears on the 1916
map (Figure 5.3) is visible
here as a depression in the
landscape. At this time, the
Mosfell farm had moved
down into the centre
of the valley (Mosfell
Archaeological Project).
doors of the farmstead suggest that the original farm at
Mosfell may have been located in the same place as the
nineteenth-century farm. This seems plausible, but the
repeated bulldozing and modern construction in this
location makes archaeological verification difficult.
Fortunately, archaeological and historical evidence
for the church and graveyard associated with the Mosfell
farmstead is much better. There appears to have been
three separate phases of the Mosfell church: 1) medieval,
twelfth–fifteenth century; 2) pre-modern, early sixteenth
century to 1888; 3) modern, 1965 to present. According
to Egils saga, the first church was built at Mosfell in the
twelfth century. This account is verified by the appear-
ance of the Mosfell church in Bishop Páll’s register of the
churches in the diocese of Skálholt in ad 1200.37 The
second church and churchyard is described by Magnús
Grímsson and the churchyard is shown on the farmstead
map drawn in 1916 (Figure 5.3), although by 1916, the
second church had already been taken down. During
the period between 1888 and 1965 when the church no
longer existed at Mosfell, the functioning farm moved
down to the wetter centre of the valley between the
Kaldakvísl and Suðurá rivers. The farmstead returned to
its current location at Mosfell in 1965 and a new church
37 Diplomatarium islandicum, ed. by Jón Sigurðsson and
others, xii, 9.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley 65

Figure 5.5.
Aerial photograph of
Mosfell from 1991,
showing the concrete
parson’s house that was
built on top of the old farm
remains seen in Figures
5.3 and 5.4. The modern
Mosfell church has been
rebuilt on almost the same
location as the nineteenth-
century church (Mosfell
Archaeological Project).

Figure 5.6.
Aerial photograph of the
modern Mosfell farm
from 2006, showing the
location of the 1995
excavations where MAP
uncovered the remains
of a medieval church
(Mosfell Archaeological
Project).

was built atop the ruins of the pre-modern church and to
the south of the medieval church. At this time, the main
farm buildings were rebuilt at the Mosfell site. The farm
buildings from the 1960s have since been levelled and a
new parish house built (see Figure 5.6).
The nineteenth-century descriptions of the ruins of
an older church lying to the north of the nineteenth-
century church provide information about the archae-
ological remnants of the medieval church. In the mid-
nineteenth century, Magnús Grímsson estimates the size
of the graveyard to be 35 × 25 m. The distinctive appear-
ance of the church wall and the east-west oriented graves
leave Grímsson with no doubt that these were the ruins
of a Christian churchyard. Of significance for locating
66
the church is Grímsson’s statement that the church and
the visible remains of the churchyard lay north of the
road leading west from the Mosfell farm. A comparison
of the aerial photographs from 1954 and 2006 (Figures
5.4 and 5.6) suggests that the location of the graveyard
ruins described by Grímsson would have been under-
neath and to the north of the current gravel road.
MAP excavations in 1995 unearthed the remains of
a medieval building that are consistent with the appear-
ance and location of the medieval church. 38 Only the
northwest corner of this building was exposed, but a
great deal was learned from this partial excavation. The
building was oriented east-west and the western wall of
the building was constructed from wooden planks set
into a wall trench. By contrast, the northern wall com-
prised a thick turf wall overlying a base of large stone cob-
bles. Landnám Tephra found in situ beneath the foun-
dations of the building indicate that it was constructed
after ad 871, while the deposition of Katla tephra from
the ad 1500 eruption atop the collapsed walls indicates
that the building pre-dates ad 1500. Unfortunately, no
contemporaneous burials were uncovered, which would
have provided the final confirmation of this building’s
function. All other indications, however, including the
38 Earle and others, Mosfell Archaeological Project 1995 Field
Season.
Davide Zori
Figure 5.7.
Aerial photograph of the
traditional Leiruvogstunga
farmstead from 2006,
showing the location of
the medieval farm. The
modern Leirvogstunga
farmstead is split into six
separate homes named
Leirvogstunga ii–vii
(Mosfell Archaeological
Project).
east-west orientation of the building, the wooden west-
ern façade, the location of the structure in the middle of
an old churchyard, and the medieval date supplied by
the tephra layers, suggest that these are the remains of
a medieval church. Overall, the archaeological evidence
from the current Mosfell farm indicates that an impor-
tant medieval farmstead with a church was located here
well before ad 1500, but there is no evidence of a farm
here in the Settlement Period. This is consistent with the
textual account of the transfer of the valley’s primary
farm to this location in the twelfth century.
The Other Primary Farms:
Leirvogstunga and Helgadalur
As suggested by the place name analysis, the farm names
Leirvogstunga and Helgadalur have the construction
and character of early primary settlement names. Leir-
vogstunga (Tidal-flats-bay-promontory), originally called
simply Tunga (Promontory), takes its name from the
prominent spit of land jutting out into the bay between the
Leirvogsá and Kaldakvísl rivers. Helgadalur (Holy-valley)
carries the name of the large subsidiary valley that extends
south of the east-west axis of the Mosfell Valley. The three
primary farms, Hrísbrú/Old Mosfell, Leirvogstunga,
and Helgadalur, are evenly distributed across the valley
landscape, maximizing the distance between the farms
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
within the productive areas of the valley (Map 5.1).
Leirvogstunga and Helgadalur may also have had spe-
cialized functions or roles in the valley’s earliest cultural
landscape. Specifically, the placement of Leirvogstunga
is oriented towards maritime resources and possibly con-
nected to the monitoring of ocean-going trade and travel,
whereas the place name evidence suggests that Helgadalur
held special ritual or religious importance.
Leirvogstunga
Leirvogstunga first appears in the historical sources
in tax records from 1547 to 1548, but the place name
Leirvogstunga/Tunga suggests that it was one of the
first farms in the valley. Archaeological rescue excava-
tions undertaken at Leirvogstunga by Fornleifastofnun
Íslands before the construction of a series of apartments
documented a farmstead of substantial size, inhabited
continuously throughout the Viking Age and Medieval
Period. 39 The excavators interpreted this farm as a
medium-sized holding of intermediary status, probably
belonging to a free landowning farmer.40 This interpre-
tation fits well with the settlement order and hierarchy
proposed in this chapter.
The centre of the pre-modern farm lies beneath
the farm mound, visible before the excavation as a cul-
tural modification of a natural rise between the mod-
ern Leirvogstunga farms ii and iii (Figure 5.7). 41
Excavations between Leirvogstunga ii and iii encoun-
tered midden deposits and the remains of numerous
non-domestic structures dating to the Viking Age and
Medieval Period. A 30 m east-west trench descending
the farm mound westwards towards the Leiruvogur Bay
encountered deep midden layers. This was likely the pri-
mary location for kitchen trash disposal, and the south-
ern profile of the trench shows cultural deposits up to
1.4 m deep. The in situ ad 1226 Medieval Tephra Layer
appears in the middle of these cultural deposits, clearly
showing a substantial occupation for a long period prior
to ad 1226.
Thirty-two structures were identified in an open-area
excavation. These excavations did not unearth the main
medieval habitation structure, which is believed to be
39 Lilja Björk Pálsdóttir and Oddgeir Hansson, Fornleifarann-
sóknir á Deiliskipulagssvæði í Leirvogstungu, Mosfellbæ.
40 Lilja Björk Pálsdóttir and Oddgeir Hansson, Fornleifarann-
sóknir á Deiliskipulagssvæði í Leirvogstungu, Mosfellbæ.
41 Elín Ósk Hreiðarsdóttir, Deiliskráning í Leirvogstungu við
Mosfellsbæ.
67
located further to the west, in an area that has not been
excavated. Nonetheless, this excavation provided docu-
mentation that is consistent with the establishment of
the Leirvogstunga farmstead soon after the colonization
of Iceland. Specifically, the presence of Landnám Tephra
in laid structural turf blocks suggests that these turf fea-
tures were built not long after the Landnám Tephra fell
in ad 871 ± 2. The most significant element of the oldest
farmstead encountered in the excavations is the remains
of the turf wall surrounding the old farm buildings. The
Medieval Tephra Layer from ad  1226 was identified
above this wall, reinforcing the pre-1226 date of its con-
struction.
Several of the excavated buildings, dating to the
Medieval Period between 1226 and 1500, provide evi-
dence for a medium to large farmstead with structures
serving specialized purposes. The prevalence of fish
bones in many of the floor layers and the recovery of at
least one net sinker indicate that at least a portion of the
farm’s economy was directed towards the exploitation of
marine resources.42
Helgadalur
Although the Helgadalur farm is first mentioned in the
textual record in ad 1395, when it is listed as belong-
ing to the Viðey monastery, place name evidence sug-
gests that Helgadalur was the first farm in the subsidi-
ary valley bearing the same name and most likely one
of the first settlements within the larger Mosfell Valley.
Archaeological testing of this farm conducted by MAP
in 2009 revealed that the Helgadalur farm was occupied
very soon after ad 870 and therefore appears to be one
of the earliest farms in the Mosfell Valley.43
The place name Helgadalur, ‘Holy Valley’ or less likely
‘Helgi’s Valley’, may contain indications of cultural mean-
ing and use in connection with the pre-Christian Norse
religion. Several place names in the vicinity include the
same prefix: Helgafell (Holy Mountain), Helgusel (Holy
Shieling), and Helgufoss (Holy Waterfall). The prefixes
in Helgusel and Helgufoss, both located in a verdant hol-
low renowned in tradition for being grass-rich and free
from the wind, could derive from the adjective holy, but
cannot be possessives of the personal name Helgi (gen.
Helga). If these names are to be interpreted as contain-
42 Data concerning these finds from Leirvogstunga excavation
from Lilja Björk Pálsdóttir and Oddgeir Hansson, Fornleifarann-
sóknir á Deiliskipulagssvæði í Leirvogstungu, Mosfellbæ, pp. 76, 79.
43 Byock and Zori, Farmstead Survey in the Mosfell Valley.
68
Figure 5.8. Map of the Helgadalur farm from 1916.
The modern farm buildings of the Helgadalur farm have
shifted slightly to the east as can be seen by comparing this
drawing with the aerial photograph in Figure 5.9 (map by
author based on 1916 original in Þjóðskálasafn Íslands).
ing possessives, they would instead be the possessive of
the female name Helga (Helgu). The most logical expla-
nation is that the names of all these neighbouring places
contain the single adjectival prefix ‘holy’, rather than the
names of two separate individuals, a man named Helgi
and a woman named Helga.
Saga references to pagan holy mountains and the dis-
tinctive shape of the Helgafell Mountain, located on the
west side of the Helgadalur Valley, suggest that the asso-
ciation of these topographical features with holiness pre-
dates Christianization around ad  1000. One famous
saga story tells of a holy mountain on the peninsula
Snæfellsnes that stood out prominently in the landscape,
and which was also given the name Helgafell. There are
in Iceland eight mountains named Helgafell.44 The ear-
liest settlers may have recognized the distinctive shape
of the Helgafell Mountain, which was conspicuously vis-
ible from the sea, and fit this mountain into their cog-
nitive framework of the appearance of holy mountains.
The secondary valley, extending behind the Helgafell
Mountain, may in this way have acquired a similar asso-
ciation with holiness.
The farm map from 1916 shows the pre-modern
Helgadalur farm located west of and partially under-
44 Svavar Sigmundsson, ‘Analogi i Islandske Stednavne’, p. 191.
Davide Zori
neath the current farm (Figures 5.8 and 5.9). Further
to the west, a ridge consisting partially of accumulated
cultural material stretches into the hay field, suggest-
ing the presence of an even earlier occupation phase.
Preliminary site survey identified midden material erod-
ing out of the stream bank adjacent to the location of
the pre-modern farm and the older cultural ridge. These
observations indicated that sub-surface testing might be
able to identify diagnostic cultural materials from the
site’s earliest occupation that could be used to evaluate
the age of the farmstead.
Systematic sub-surface testing, consisting of coring
transects following the long axis of the ridge, confirmed
that it had been formed through accumulation of cul-
tural material. 45 Layers containing ash, charcoal, and
burnt bone were found to a depth of 50 cm below the
surface. A compacted outside surface area upon which
farmstead activities took place was recognized in cores
on the ridge, suggesting that an old farm would have been
close by. Most significantly, the testing demonstrated
human settlement at the Helgadalur farm very soon after
the Landnám Tephra fell in ad 871 ± 2. Charcoal of
cultural origin within an orangish-gray anthropogenic
deposit was found immediately above a lens of in situ
Landnám Tephra. Small charcoal fragments immediately
above the Landnám Tephra in one of the modern drain-
age trenches to the northwest of the ridge provide addi-
tional evidence for the substantial nature of this early
occupation.
Examination of the eroding slope of the streambank
south of the cultural ridge revealed deposits from the
medieval occupation of Helgadalur, including mixed
midden with charcoal, bone, peat ash, and re-deposited
turf. Coring through the slope of the profile revealed
stratified turf blocks containing multiple Landnám
Tephra layers and a well-defined layer of thin black
tephra, probably Katla-R from c. 920 or Eldgjá-1 from
c. 934. Despite some disturbance and post-depositional
mixing by the on-going erosion of the stream bank, the
inclusion of substantial amounts of Landnám Tephra
in the turf blocks suggests a Viking Age (settlement to
ad 1100) date for portions of this mixed deposit.
45 For this and other sub-surface coring sur veys (e.g.
Skeggjastaðir) we employed a coring device consisting of a dry
sampling tube with a 1.25 inch diameter (3.18 cm) and a length
of 18 inches (45.72 cm). A telescoping pole on the coring device
allowed the core to reach a depth of just over 110 cm.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
Secondary -staðir Farms
Three farms in the Mosfell Valley, Skeggjastaðir, Hraða-
staðir, and Æsustaðir, contain the habitative place name
element -staðir, suggesting that they were secondary
farms dating to the early period soon after the initial
settlement of the valley. The place name model suggests
that the Icelandic -staðir farms were founded as splinter
farms from primary settlements. Following this model,
the -staðir farms in the Mosfell Valley should be subsidi-
ary foundations from one of the three primary farms in
the valley, most probably from Mosfell, which according
to the place name evidence, the texts, and the archaeo-
logical evidence, was the dominant farm in the valley.
The names of the three secondary farms all have a
personal name as the first element and therefore do not
help us understand the reasons for the establishment of
the farms. The three secondary -staðir farms distribute
evenly across the valley landscape, filling available loca-
tions between the earlier primary farms (Map 5.2). The
locations of these farms therefore appear to be predi-
cated on the availability of grazing land and the maximi-
zation of settlement spacing on the landscape.
Skeggjastaðir
The Skeggjastaðir farmstead is of particular interest
historically, as the recorded farmstead of the original
69
Figure 5.9.
Aerial photograph of the
Helgadalur farm from
1997 with the Settle-
ment Period farmstead
remains marked (Mosfell
Archaeological Project).
settler in the area, and also methodologically, because
of the apparent disagreement between the textual and
place name traditions. According to Landnámabók,
Þórðr Skeggi, the first settler in the Mosfell region, set-
tled at Skeggjastaðir in the far eastern end of the Mosfell
Valley.46 Some scholars have doubted whether the first
settler would have chosen to establish his primary farm
in this location,47 since it is relatively far from the ocean
and appears to have several environmental disadvan-
tages. For example, Skeggjastaðir lacks wetland meadow
grazing lands and is located in a basin that collects snow
earlier in the winter and retains it later into the spring
than sites lower in the valley. On the other hand, other
environmental factors may have been attractive for
early settlers, such as the proximity to a river known for
salmon, the well-drained grassy slopes leading down to
the river, and easy access to highland grasslands in a val-
ley initially covered with birch trees.48 A relevant passage
from Landnámabók highlights the role of human agency
and individual preferences in the settlement process,
‘Sumir þeir, er fyrstir kómu út, byggðu næstir fjöllum ok
merkðu at því landskosina, at kvikféit fýstisk frá sjónum
46 Íslendingabók, Landnámabók, ed. by Jakob Benediktsson.
47 Magnús Grímsson, ‘Athugasemdir við Egils sögu Skallagríms-
sonar’.
48 Bjarki Bjarnason and Magnús Guðmundsson, Mosfellsbær.
70 Davide Zori

Figure 5.10.
Map of Skeggjastaðir,
showing the location
of the medieval farm-
stead discovered with
sub-surface coring.
The remains of the
old homefield wall
that surrounded the
protected hayfield are
still visible and shown
on the map. These
remains correspond
to the homefield wall
drawn on the 1916
map of the farm
(Figure 5.11)
(Mosfell Archaeo-
logical Project).

til fjallana’ (‘some of those who came first to Iceland set-

Figure 5.11. Map of Skeggjastaðir farmstead from 1916.
The farmhouses are located north of the farm access road while a
garden divided in two appears south of the road. The homefield wall
is shown as the outer dark line surrounding the farm (map by author
based on 1916 original in Þjóðskálasafn Íslands).
tled in the mountains and concerning the quality of the
land they observed that the livestock grazed from the sea
to the mountains’).49
Systematic coring at Skeggjastaðir conducted by
MAP in 2009 discovered the site of a medieval farm-
stead that dates to the Viking Age (settlement to
ad 1100), if not more narrowly to the Settlement Period
(ad 870–930; Figure 5.10).50 The farm is situated near
the edge of a steep gully through which a small tributary
stream flows from south to north down to the Leirvogsá
River. This placement is ideal for access to fresh water,
while the location high on the slope offers a good view of
the surrounding landscape.
An important feature in the landscape that aided the
2009 archaeological survey at Skeggjastaðir was a sub-
stantial old homefield wall (túngarður), which may date
back to the Medieval Period (Figure 5.10). This wall,
which appears in the same shape on the 1916 farmstead
map, delimited the extent of the protected hay fields
of the traditional farm (Figure 5.11). Although previ-
49 Íslendingabók, Landnámabók, ed. by Jakob Benediktsson,
p. 337. The translation is my own.
50 Byock and Zori, Farmstead Survey in the Mosfell Valley.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
ous surveys had not identified the early farm, the 2009
programme of systematic coring on a 20 × 20 m grid
within the old homefield successfully located the medi-
eval farm. After identification of the farmstead, the grid
was tightened to 10 m intervals between each core to
increase the resolution of the archaeological deposits
(Figure 5.12).
The remains of the farm documented through coring
stretched across an area measuring approximately 40 m
N-S and 20 m E-W (E 1100–20 and N 960–1000). The
sub-surface survey recovered evidence of a large occu-
pation area, including remains of turf walls, peat ash
midden layers, and an extensive outside living surface
surrounding the farm buildings (Figure 5.13). The medi-
eval date of these farmstead remains is demonstrated
by the Katla Tephra from ad 1500, which lies atop the
occupation layers in many of the core samples. The turf
walls observed in cores SK 77, 78, 79, and 80 contained
Landnám Tephra (ad 871 ± 2), suggesting an early post-
settlement date for the construction of the walls. The
turf walls and at least the lowest midden and occupation
layers in the area date to the same period, probably not
long after the Landnám Tephra fell.
In addition to the temporally diagnostic tephra lay-
ers, coring at Skeggjastaðir revealed an outside living sur-
face with the remains of concentrated domestic activity.
71
Figure 5.12.
Skeggjastaðir map with sub-surface
coring numbers shown. Evidence of the
early farm at Skeggjastaðir consisted of
turf walls containing Landnám Tephra
(ad 871 ± 2), midden layers dumped
down the slope towards the small stream,
and a compacted outside living surface
upon which farm activities took place
(Mosfell Archaeological Project).
This outside surface area is characterized by horizontally
compacted sediment, high concentrations of charcoal,
and linear red oxidation stains created by small puddles
and low-grade iron-panning. The unique character of
this layer, upon which the medieval inhabitants walked
and lived, was created by the compacting and sorting
effects of human ambulation. This layer is similar to the
outside surface area identified around the longhouse at
Hrísbrú/Mosfell and in the cores from the Helgadalur
survey. This distinctive layer appeared in cores SK 71, 72,
73, 74, and 75, primarily north and east of the cores that
contained turf walls, possibly indicating that outside
work took place to the north and east of the domestic
structures. Stratigraphic superposition of the Katla 1500
tephra over the outside living surface in core SK 73 indi-
cates that this surface area dates to the Medieval Period
(Figure 5.13).
Midden layers were identified in several core sam-
ples surrounding the medieval structure, and comprised
dense layers of ash, charcoal, and food remains.51 Based
on their relative depth in the soil and their proximity
to the early turf walls and the pre-ad 1500 occupation
area, these midden layers are most likely associated with
51 Midden was identified in cores SK 61, 64, 77, and 79.
72
the same medieval farmstead. The thickest midden layers
appeared in cores close to the edge of the ravine leading
down to the stream (Figures 5.12 and 5.13). The inhab-
itants of the suggested domestic structures located just
to the east probably dumped kitchen waste down the
nearby slope and into the stream, a practice also docu-
mented at the medieval farm at Helgadalur.
The archaeological investigations at Skeggjastaðir
identified an early occupation that is consistent with
both the textual record and the place name model. The
nature of the archaeological evidence, however, means
that it could not clarify the discrepancy between the
Landnámabók textual record that identifies this as the
first settlement in the valley and the place name tradi-
tion that suggests that Skeggjastaðir was a secondary set-
tlement. The discrepancy in settlement time would most
likely be too small to be distinguishable archaeologically.
Instead, the difference between primary and secondary
sites should be sought in the comparison of the site size
and status, as indicated by the archaeological remains.
Although too little evidence has yet been recovered from
Skeggjastaðir, the chiefly establishment at the Hrísbrú/
Davide Zori
Figure 5.13.
Section drawings of cores from
the medieval farmstead at
Skeggjastaðir. Note the turf wall
with Landnám Tephra in core
SK 79, the midden in SK 61,
and the outside living surface in
SK 73 that appears below the
Katla tephra from ad 1500.
Old Mosfell suggests that it ranks highest in the regional
settlement hierarchy, supporting the place name model.
In this case, then, the textual record appears to present
an instance where the memory of the initial settlement
site was lost and people subsequently rationalized that
the original farm in the valley would have been named
after the first settler, Skeggi.
Hraðastaðir
The first textual reference to the Hraðastaðir farm appears
in the Viðey monastery charter from ad 1395.52 According
to oral tradition, however, the farm is one of the oldest in
the valley,53 and the place name suggests that the farm has
an early origin as a subsidiary farm to a primary farm in
the region, presumably Hrísbrú/Old Mosfell.
52 Diplomatarium islandicum, ed. by Jón Sigurðsson and
others, iii.
53 Magnús Grímsson, ‘Athugasemdir við Egils sögu Skallagríms-
sonar’.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
Due to the increasing pace of construction and devel-
opment in the Mosfell Valley, archaeological research
in the valley is increasingly difficult at some of the tra-
ditional farms. Hraðastaðir is an example of a farm
where development has rendered impractical research-
driven archaeological sub-surface survey and excavation.
Extensive modern construction has taken place on the
site, resulting in nine private homes and associated build-
ings and features on the property of the old farmstead
(Figure 5.14). The modern farm named Hraðastaðir I is
located on top of the nineteenth-century farm and pos-
sibly also atop the remains of the medieval farm. Even
limited sub-surface survey was deemed too hazardous,
as anthrax (Bacillus anthracis), probably from diseased
horses from around 1900, was identified in the ground
northeast and east of the modern farmhouse. The ruins
of just one structure were tested, east of the old home-
field. Sub-surface testing showed that the structural
ruins are a post-medieval animal shed with hearth trash
dumped into the building before it collapsed. The turf
walls of the building contain Katla 1500 tephra, indicat-
ing its construction after that date.
In the case of Hraðastaðir, sub-surface survey was
unable to help in establishing the antiquity of the farm.
A closer look at the intersection of the historical land-
scape, oral traditions, and the early texts, however, has
greater potential. Landnámabók does not directly men-
73
Figure 5.14.
Aerial photograph of
Hraðastaðir from 2006,
showing the location
of the old farmhouse
and the post-medieval
ruins of a likely animal
shed (HRÐ-1).
Significant modern home
construction at this site
makes archaeological
work difficult. Midden
deposits dumped into the
abandoned ruin HRÐ-
1 might indicate that
the older pre-modern
and medieval farm was
located closer to this ruin
than the current farm
(Mosfell Archaeological
Project).
tion the farm of Hraðastaðir, but it does mention a man
named Þorbjörn Hraðason from Mosfell. If this tradition
is true, then it might imply that the farm was founded
by Þorbjörn’s father, Hraði. A local oral tradition dating
back to at least the end of the eighteenth century may
corroborate the antiquity of the Hraðastaðir farmstead.
This oral tradition holds that a low mound at the bot-
tom of the valley is called Hraðaleiði or ‘Hraði’s burial
mound’.54 The Hraðaleiði mound, which is 1.7 m high,
10–12 m long and 6 m wide, is distinct in the landscape,
with no other mounds in the surrounding area. Erosion
of the mound has exposed loose and gravelly sediment
that appears to have been redeposited.
The Hraðaleiði mound is particularly interesting in
light of its location at the intersection of the traditional
boundaries of the Mosfell, Hraðastaðir, and Æsustaðir
farms as recorded in 1817 (Figure 5.15).55 The location
of the mound is consistent with criteria for the placement
of Icelandic pagan burial mounds: on the boundaries of
farms, outside of cultivated fields, and in a prominent
54 Sveinbjörn Rafnsson, ‘Fornleifaskýrsla Sera Markúsar
Sigurðssonar á Mosfelli’, pp.  258–59. Agnes Stefánsdóttir and
others, Skráning Fornleifa í Mosfellsbæ, p. 57.
55 Sveinbjörn Rafnsson, ‘Fornleifaskýrsla Sera Markúsar
Sigurðssonar á Mosfelli’, p. 258
74
place in the landscape.56 The correspondence of farm
boundaries and the location of the Hraðaleiði mound
is not likely to be an accident, meaning that it is either
a genuine pagan burial or a very old oral tradition that
was co-opted into the land division of the Mosfell Valley.
The Hraðaleiði mound, as a relic of the process of territo-
rial marking in the pagan period, suggests that the farm
boundaries recorded in the early nineteenth century
may at least in part correspond with the original farm
boundaries established during the early settlement of the
valley before the conversion to Christianity in ad 1000.
This evidence suggests that the three farms sharing this
boundary intersect – Hrísbrú/Mosfell, Hraðastaðir and
Æsustaðir – were established prior to ad 1000.
Æsustaðir
The first textual reference to the Æsustaðir farm appears
in 1704 in Magnússon and Vídalín’s Jarðabók.57 The
place name with the -staðir suffix, however, suggests
that the farm has a much earlier origin as a second-
ary farm in the valley. The place name Æsustaðir con-
tains the feminine personal name Æsa, a reminder that
56 Adolf Fríðriksson, ‘Social and Symbolic Landscapes in Late
Iron Age Iceland’, pp. 9–21.
57 Jarðabók Árna Magnússonar og Páls Vídalíns, pp. 318–19.
Davide Zori
Figure 5.15.
Aerial photograph
showing the location of
the Hraðaleiði mound
at the intersection of the
traditional boundaries of
the Mosfell, Hraðastaðir,
and Æsustaðir farms.
The placement is consis-
tent with criteria for the
placement of Icelandic
pagan burial mounds on
the boundaries of farms,
outside of cultivated
fields, and in prominent
places in the landscape
(Mosfell Archaeological
Project).
Norse society allowed women to claim, own, and man-
age farmsteads. An oral tradition concerning a pagan
burial mound called Æsuleiði, in which the presumed
owner of the farm, Æsa, had been laid to rest, provides
hints that the farmstead predates the island’s conversion
to Christianity (Figure 5.16). Magnús Grímsson sug-
gested, without specifying his evidence, that Æsustaðir
was the original farm occupying the northern slopes
of the Helgafell and Æsustaðafjall mountains and that
this farm was later split into three: Æsustaðir, Norður
Reykir, and Hlaðgerðarkot.58
The place name model also suggests Æsustaðir was
the first farm in the southern half of the Mosfell Valley,
probably utilizing the grazing land stretching across
the northern slopes of the Æsustaðafjall and Helgafell
mountains. The topographic appellative of the mountain
to the south, Æsustaðafjall (‘Æsustaðir’s-mountain’),
derives from the name of the farm. Names of large geo-
graphic features such as mountains are generally con-
servative, making it unlikely that Æsustaðafjall had a
different name originally. Æsustaðir therefore should
predate the other two farms on this mountain’s slopes,
Norður-Reykir and Hlaðgerðarkot. If either of these
58 Magnús Grímsson, ‘Athug asemdir við Eg ils sög u
Skallagrímssonar’, p. 263.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley 75

Figure 5.16. The Æsustaðir farm in 1954. According to oral tradition the female founder of Æsustaðir was interred in a pagan resting place
called Æsuleiði (Æsa’s burial mound). One current tradition holds that Æsuleiði is a small mound just outside the old homefield.
Excavation showed, however, that this is not a burial mound, but a mound of turf, hay, and manure created after ad 1500.
The age of the link between the oral tradition and the place name is unknown, but the archaeological demonstration that the link
post-dates ad 1500 shows the post-medieval concern with identifying such monuments in the landscape. The neighbouring farm
Norður Reykir, seen on the left, is a tertiary farm founded later in the Middle Ages (Mosfell Archaeological Project).

farms were established before Æsustaðir, the mountain
should be named after one of these two farms.
Preliminary sub-surface archaeological investigations
at Æsustaðir showed significant anthropogenic soils below
the Katla 1500 tephra, indicating that the farm was in use
in the Medieval Period. Although vague, this archaeologi-
cal evidence provides a date for the farm that is over two
hundred years earlier than that given in the textual records
and consistent with the place name model predicting an
early settlement period date for the Æsustaðir farm.
Tertiary Farms
Based on the character of their place names, the remain-
ing pre-modern farms in the Mosfell Valley are classi-
fied as tertiary farms (Map 5.3). Their names and the
historical records concerning these farms are consistent
in suggesting that they were established as part of the
gradual agricultural intensification leading to the divi-
sion of the lands of the primary and secondary farms in
the Medieval or early Post-Medieval Period. Extensive
archaeological excavation has taken place at the current
Hrísbrú farm, but little archaeological work has been
conducted at the other sites.
Hrísbrú
Hrísbrú, ‘Reed-bridge or causeway’, is most likely the
oldest of these tertiary farms. Egils saga includes a ref-
erence to the establishment of a farm by this name in
c. ad 1150, at the place where the Mosfell farmstead was
previously situated. The story of the relocation of the
primary Mosfell farm and the foundation of a subsidi-
ary splinter farm at Hrísbrú is consistent with the gen-
76
eral progression of the settlement of the Mosfell Valley
proposed in the place name settlement model. The
archaeological evidence of the large Viking Age farm at
Hrísbrú was addressed above in the section on Mosfell
because of the relocation of the farm and concomitant
shift in the place name. The archaeological data from
Hrísbrú dating from the middle of the twelfth century to
the modern period includes a disturbed post-medieval
agricultural structure and a significant quantity of small
finds, including iron fragments and nails, and redware
and whiteware glazed ceramics that would have been
common at all farms. These finds do not tell us much,
but are consistent with the textual record of continued
habitation at the site. One interesting note regarding the
post-medieval farm at Hrísbrú is the continued memory
and reverence of the old churchyard, as illustrated by the
burial of a single male individual in the long-abandoned
graveyard during the fourteenth century.59
Laxnes
Laxnes, ‘Salmon-promontory’, appears in the Danish
king’s Fógetareikningar tax records from 1547–52.
Laxnes is a natural topographical name referring to
a small tongue of land between the Kaldakvísl River
and a small tributary stream. In the landscape, this
tongue is not a prominent topographical feature such
as the Helgadalur Valley, the Mosfell Mountain, or the
Leirvogstunga Promontory. Several factors indicate that
Laxnes should be classified as a tertiary farm: 1) it takes
its name from a minor topographical feature, 2) the place
name contains no habitative appellatives such as -staðir
to provide clues for an earlier settlement, 3)  the farm
is absent from textual records until the Post-Medieval
Period, and 4) the location of the farm in the centre of
the valley without the benefits of the well-drained grassy
slopes of hills or mountains.
Minna-Mosfell
Minna-Mosfell, ‘Little-Mosfell’, first appears in the
textual record in the Fógetareikningar document from
1547–52. This derivative name references the Mosfell
farm and adds the diminutive first element, showing
Minna-Mosfell’s subsidiary role and indicating that
59 This burial is Feature 18 in Plate 4 and Fig. 1.10 in Chapter
1 by Byock and Zori. The radiocarbon date range for this individual
is ad 1320–50. See Chapter 7 by Eng for an analysis of the skeletal
material in the Hrísbrú graveyard.
Davide Zori
it was established on land divided from the original
Mosfell lands. There are no indications from the place
name evidence or from the textual record that suggest
this tertiary farm was settled before the Late Medieval
Period.
Norður Reykir
Norður Reykir, ‘North Smoke or Steam’, is first recorded
in a land register from 1695. This name is also derivative
and takes its name from the farm Reykir, located south
of the Mosfell Valley. Both Reykir and Norður Reykir
are located next to concentrations of natural geothermal
hot springs.
Hlaðgerðarkot
Hlaðgerðarkot, an abandoned farm that occupied a
small property west of Norður Reykir, is first mentioned
in the textual record in the 1704 Jarðabók. The name
Hlaðgerðarkot, ‘Hlaðgerður’s Cottage’, is a habitative
including the final generic element -kot (‘cottage, hut,
or small farm’) that refers to a small attached settlement.
The -kot element first appears in Iceland in the four-
teenth century, after which it became popular in names
of attached tenant farms.60 This -kot generic is also used
in the small seventeenth- to eighteenth-century tenant
farm Bakkakot owned by the Mosfell farm. The first
element in Hlaðgerðarkot is the genitive of the female
name Hlaðgerður, making this the second settlement in
the valley named after a female occupant or owner.
Discussion and Conclusions
Archaeological surveys and excavations indicate
that Norse settlers established at least four farms –
Hrísbrú/Old Mosfell, Leirvogstunga, Helgadalur, and
Skeggjastaðir – during the early settlement history in the
Mosfell Valley. This is three more than the number of
farms indicated by the texts. The archaeology agrees with
the place name model that predicted that Mosfell was
the first farm in the valley, and that settlers founded sub-
sequent early settlements at Leirvogstunga, Helgadalur,
and Skeggjastaðir. The archaeology also supports the
view from the textual sources that the chiefly establish-
ment and the farm name was moved from the current
60 Svavar Sigmundsson, ‘Place-Names in Iceland and Shetland’,
p. 213.
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley
Hrísbrú site to present-day location of the Mosfell farm
during the twelfth century.
Approaching the distribution of the earliest farms
from the perspective of the landscape and geography of
the Mosfell Valley suggests an economic logic in their
placement, one that may reflect chiefly design in set-
tling supporters in specific areas to take advantage of
local resources. The paramount Hrísbrú/Old Mosfell
site is ideal for the balanced economy of a large Norse
farmstead. This site is located on the northern side of the
valley, which received the most sunshine for the great-
est part of the year. Because of this, the sunny northern
slopes of the Mosfell Mountain support the widest dry
hay meadow in the valley. The Hrísbrú-based chieftains
would have used these meadows to produce hay for
the winter, which was often the limiting factor for the
numbers of livestock that could be sustained. The farm
is positioned prominently at the mouth of the valley,
maximizing access to low wetlands ideal for grazing cat-
tle, which was an economic preoccupation of high status
households in Norse society.61
The outlying farms are characterized by a greater de-
gree of economic specialization. The location of Leir-
vogstunga and the material remains from the recent
excavations indicate an orientation toward the sea.
Skegg jastaðir is positioned ideally to exploit the riverine
resources of Leirvogsá River, as well as the grazing po-
tential of the low highlands. The Helgadalur farm fills in
the final large gap in the landscape by taking advantage
of the slopes of the subsidiary Holy Valley to the south.
In terms of the social mechanisms of settlement,
the case of the Mosfell Valley appears to support the
‘Skallagrímr type’ model of settlement. Although the
agency in the placement of the farms is difficult to deter-
mine, the evidence fits a model whereby the farm loca-
tions resulted from the chiefly design of the primary
settler. In this model, the primary settler founded the
dominant site of Mosfell at the current Hrísbrú farm
and subsequently established supporters at economically
intensifiable locations on the landscape. Later, more
peripheral farms were established to exploit specific eco-
nomic niches, such as the marine and riverine resources.
The legacy of this chieftain-led settlement process
also reflects the social stratification occurring through-
out Iceland from the ninth to the thirteenth century.
Archaeological survey indicates that the number of
farms in the valley increased gradually from the initial
61 Orri Vésteinsson, McGovern, and Keller, ‘Enduring Impacts’.
See also Zori and others, ‘Feasting in Viking Age Iceland’.
77
three large primary farms, to six before the end of the
tenth century, and eight or nine by the middle of the
thirteenth century. The stratification of these farms, as
suggested by the place names, is supported by the textual
focus on the chiefly farm at Mosfell and by the archaeo-
logical evidence of wealth and status recovered during
the excavation of the Hrísbrú/Mosfell farmstead.
Significantly, the sociological picture derived from
the textual, place name and archaeological evidence
was only very rarely contradictory. The discrepancy in
the identification of the first settlement site in the val-
ley provides the clearest example. I argue, however,
that this contradiction actually yields a much richer
story. The Landnámabók text holds that the first set-
tler to the valley, Þórðr Skeggi, established his farm at
Skeggjastaðir. Place name evidence, on the other hand,
suggests the Mosfell farm was the first settlement in the
valley and that the farm names with -staðir suffixes, like
Skeggjastaðir, were secondary settlements. The archaeo-
logical record showed Hrísbrú to be the most likely can-
didate for the primary site in the valley with the excava-
tion of a high status farmstead and the earliest datable
evidence of human habitation. The Landnámabók tra-
dition appears to have lost the memory of the first set-
tler’s farmstead. To compensate for this loss, the creators
of this tradition reconstructed – or reimagined – the
settlement history based on the place name of the farm
Skeggjastaðir, as named after the imputed first settler.
The ability to cross-check and modify imprecision
derived from the oral traditions that inform the sagas is
one of the chief strengths of interdisciplinary research
on medieval Iceland. Only reference back to the textual
record, however, can explain the incongruity between
the archaeological findings and the place name evidence.
In the case of the Mosfell Valley, the medieval texts indi-
cate that the place name Mosfell shifted from the loca-
tion of the Hrísbrú farm to the current position of the
Mosfell farm in the twelfth century. It is only through
the interdisciplinary analysis of all available sources that
we can provide the most complete, nuanced, and histori-
cally satisfying story.
78 Davide Zori

Bibliography
Archival Documents
Reykjavík, Þjóðskálasafn Íslands, ‘Helgadalur túnakort’ (unpublished map, 1916)
—— , ‘Hraðastaðir túnakort’ (unpublished map, 1916)
—— , ‘Hrísbrú túnakort’ (unpublished map, 1916)
—— , ‘Leirvogstunga túnakort’ (unpublished map, 1916)
—— , ‘Mosfell túnakort’ (unpublished map, 1916)
—— , ‘Skeggjastaðir túnakort’ (unpublished map, 1916)
—— , ‘Æsustaðir túnakort’ (unpublished map, 1916)

Primary Sources
Diplomatarium islandicum: Íslenzk fornbréfasafn, ed. by Jón Sigurðsson and others, 16 vols (København and Reykjavík: Möller and
Hið íslenzka bókmentafélag, 1857–1972)
Egils saga Skalla-Grímssonar, ed. by Sigurður Nordal, Íslenzk fornrit, ii (Reykjavík: Hið íslenzka fornritafélag, 1933)
Íslendingabók, Landnámabók, ed. by Jakob Benediktsson, Íslenzk fornrit, i (Reykjavík: Hið íslenzka fornritafélag, 1968)
Jarðabók Árna Magnússonar og Páls Vídalíns, ed. by Árni Magnússon and Páll Víðalín, 13 vols (København: Hið íslenska fræðafjelag,
1913–43)

Secondary Studies
Adolf Fríðriksson, ‘Social and Symbolic Landscapes in Late Iron Age Iceland’, Archaeologia Islandica, 7 (2009), 9–21
Adolf Friðriksson and Orri Vésteinsson, ‘Creating a Past: A Historiography of the Settlement of Iceland’, in Contact, Continuity,
and Collapse: The Norse Colonization of the North Atlantic, ed. by James Barrett, Studies in the Early Middle Ages, 5 (Turnhout:
Brepols, 2003), pp. 139–61
Agnes Stefánsdóttir, Rúna Tetzschner, Guðmundur Ólafsson, Ágúst Georgsson, Kristinn Magnússon, and Bjarni Einarsson, Skráning
Fornleifa í Mosfellsbæ (Reykjavík: Þjóðminjasafn Íslands, 2006)
Andersen, Per Sveaas, ‘The Norwegian Background’, in Scandinavian Settlement in Northern Britain: Thirteen Studies of Place-Names
in Their Historical Context, ed. by Barbara Crawford (London: Leicester University Press, 1995), pp. 16–25
Árni Daníel Júlíusson, Jón Ólafur Ísberg, and Helgi Skúli Kjartansson, eds, Íslenskur sögu atlas 1: Frá öndverðu til 18. aldar (Reykjavík:
Iðunn, 1991)
Bjarki Bjarnason and Magnús Guðmundsson, Mosfellsbær: Saga Byggðar í 1100 Ár (Reykjavík: Pjaxi, 2005)
Bolender, Douglas, John Steinberg, and Brian Damiata, ‘Farmstead Reorganization at the End of the Viking Age: Results of the
Skagafjörður Archaeological Settlement Survey’, Archaeologia Islandica, 9 (2011), 77–101
Bühnen, Stephen, ‘Place Names as an Historical Source: An Introduction with Examples from Southern Senegambia and Germany’,
History in Africa, 19 (1992), 45–101
Byock, Jesse, and Davide Zori, Farmstead Survey in the Mosfell Valley: Mosfell Archaeological Project, 2009 (Reykjavík: Unpublished
report submitted to Fornleifavernd Ríkisins, 2009)
Byock, Jesse, Phillip Walker, Jon Erlandson, Per Holck, Davide Zori, Magnús Guðmundsson, and Mark Tveskov., ‘A Viking-Age
Valley in Iceland: The Mosfell Archaeological Project’, Medieval Archaeology, 49 (2005), 195–218
Byock, Jesse, Phillip Walker, Jon Erlandson and Per Holck, Viking Age Graves in the Mosfell Valley: The Mosfell Archaeological Project,
2001 (Reykjavík: Unpublished report submitted to Fornleifavernd Ríkisins, 2001)
Byock, Jesse, Viking Age Iceland (London: Penguin, 2001)
Callmer, Johan, Trade Beads and Bead Trade in Scandinavia, ca. 800–1000 ad (Bonn: Habelt, 1977)
Earle, Timothy, Jesse Byock, Phillip Walker, and Sigurður Bergsteinsson, Mosfell Archaeological Project 1995 Field Season (Reykjavík:
Unpublished report submitted to Fornleifavernd Ríkisins, 1995)
Elín Ósk Hreiðarsdóttir, ‘Appendix A: The Glass Beads from Hrísbrú 2008’, in Farmstead Survey in the Mosfell Valley: Mosfell
Archaeological Project, 2009, ed. by Jesse Byock and Davide Zori (Reykjavík: Unpublished report submitted to Fornleifavernd
Ríkisins, 2009)
—— , Deiliskráning í Leirvogstungu við Mosfellsbæ (Reykjavík: Unpublished report from Fornleifastofnun Íslands (FS321-06231),
2006)
Gammeltoft, Peder, The Place-Name Element Bólstaðr in the North Atlantic Area (København: Reitzel, 2001)
Gunnar Karlsson, The History of Iceland (Minneapolis: University of Minneapolis Press, 2000)
Jón Jóhannesson, A History of the Old Icelandic Commonwealth (Winnipeg: University of Manitoba Press, 1974)
Interdisciplinary Modelling of Viking Age and Medieval Settlement in the Mosfell Valley 79

Kruse, Arne, ‘Norse Topographical Settlement Names on the Western Littoral of Scotland’, in Scandinavia and Europe 800–1350:
Contact, Conflict, and Coexistence, ed. by Jonathan Adams and Katherine Holman, Medieval Texts and Cultures of Northern
Europe, 4 (Turnhout: Brepols, 2004), pp. 97–107
Kålund, Kristian, Bidrag til en Historisk Topografisk Beskrivelse af Island: i. Syd- og Vest-Fjærdingerne (København: Gyldendalsk, 1877)
Lilja Björk Pálsdóttir and Oddgeir Hansson, Fornleifarannsóknir á Deiliskipulagssvæði í Leirvogstungu, Mosfellbæ: Framvinduskýrskla
iii (Reykjavík: Unpublished report by Fornleifastofnun Íslands (FS368-06353), 2008)
Magnús Grímsson, ‘Athugasemdir við Egils sögu Skallagrímssonar’, in Safn til sögu Íslands og íslenzkra bókmennta að fornu og nýju ii,
ed. by Jón Þorkelsson (København: Hið íslenzka bókmenntafélag, 1886 [1861]), pp. 251–76
Magnús Sigurgeirson, ‘Appendix C: Archaeological Investigation at Hrísbrú in Mosfelldalur, SW-Iceland, Tephrochronoloical Study’,
in Viking Age Graves in the Mosfell Valley: The Mosfell Archaeological Project, 2001, ed. by Jesse Byock and others (Reykjavík:
unpublished report submitted to Fornleifavernd Ríkisins, 2001), section 7
Olsen, Magnus, Hvad Våre Stedsnavn Lære Oss (Oslo: Universitetet i Oslo, 1971)
Orri Vésteinsson, Thomas McGovern, and Christian Keller, ‘Enduring Impacts: Social and Environmental Aspects of Viking Age
Settlement in Iceland and Greenland’, Archaeologia Islandica, 2 (2002), 98–136
Svavar Sigmundsson, ‘Analogi i Islandske Stednavne’, in Analogi i Navngivning: Tenth Nordic Congress of Onomastics, Brandbjerg
20–24 May 1989, ed. by Gorden Albøge, Eva Meldgaard, and Lis Weise (Uppsala: NORNA, 1991), pp. 189–97
—— , ‘Íslensku staða-nöfnin’, Íslensk Mál og Almenn Málfræði, 1 (1979), 238–48
—— , ‘Place-Names in Iceland and Shetland: A Comparison’, in Cultural Contacts in the North Atlantic Region: The Evidence of
Names, ed. by Peder Gammeltoft, Carole Hough, and Doreen Waugh (Lerwick: NORNA, Scottish Place-Name Society and
Society for Name Studies in Britain and Ireland, 2005), pp. 209–15
Sveinbjörn Rafnsson, ‘Fornleifaskýrsla Sera Markúsar Sigurðssonar á Mosfelli, Dag. 1. Ágúst 1817’, in Frásögur um Fornaldarleifar
1817–1823, ed. by Sveinbjörn Rafnsson (Reykjavík: Rit Stofnun Árna Magnússonar, 1983), pp. 258–59
Zori, Davide, ‘From Viking Chiefdoms to Medieval State in Iceland: The Evolution of Power Structures in the Mosfell Valley’ (un-
published doctoral thesis, University of California, Los Angeles, 2010)
Zori, Davide, Jesse Byock, Egill Erlendsson, Steve Martin, Thomas Wake, and Kevin J. Edwards, ‘Feasting in Viking Age Iceland:
Sustaining a Chiefly Political Economy in a Marginal Environment’, Antiquity, 87 (2013), 150–65