200
into the academic mainstream in the last ten
years, with articles by Jens Peter Schjødt
(2009; 2012) and Andreas Nordberg (2012) in
particular prompting conscious debate about
unity – or lack thereof – in pre-Christian
Nordic religion. The aim of this project is to
engage with this debate by examining evidence
for variation in the locations of religious praxis
and sacrally-charged space in the Germanic
Nordic cultural area during the Late Iron Age
(ca. 500–1200 AD). The dissertation consists
of a portfolio of six articles that employ
different approaches to a range of source
material concerning a variety of religious
contexts. While each article addresses its own
independent research questions, a number of
running themes and issues underlie the
dissertation as a whole: the tension between
unity and diversity, spatialisation (the
imbuement of objective area with subjective
value), variety in the social settings of and
diachronic development in sacral space, and
the idea that distinct articulations of pre-
Christian Nordic religion can be meaningfully
identified along different axes.
The first article, “Reasoning Our Way to
Privacy: Towards
a Methodological Discourse
of Viking Studies”, is an analysis of
methodological
approaches
commonly
employed in studies of historical cultural
phenomena, including religion. It aims to
contribute to the nascent methodological
discourse of the emerging “Viking Studies”
field, and outlines and categorises a number of
methodological approaches into two loose
family groupings of “bottom-up” and “top-
down” methods. The former are described as
attempts to deduce the cultural categories
employed by bearers of a culture, that is, as
examples of Max Weber’s
Idealtypen that
depart from the concrete and attempt to
generalise on that basis (Weber 1904; cf. Frank
1997; Hall 2009: 1–20). An emic case study of
heimolleikr, a medieval Nordic cultural
concept akin to modern Western notions of
privacy, is conducted on the basis of
philological
evidence
from
medieval
manuscripts. Top-down methods are then
argued to employ inductive reasoning in their
application of external categories to cultural
phenomena (e.g. Cole 2015; Shay 2003; cf.
Fitzgerald 1997; Jensen 2003), and their results
described as examples of Ferdinand Tönnies’
Normaltypen, which move from an abstract
idea to concrete data (Tönnies 1931; 1979). An
example of etic methods is presented in the
form of a case study of “Old Norse Privacy”,
which is then compared to
heimolleikr: it is
proposed that the latter
reflects greater stress
on interpersonal relationships than is typically
the case with privacy, which features a greater
concern with the control of access to space.
Finally, the polarised nature of the presentation
of approaches in the article is once again
stressed, and the abductive reality of most
scholarship noted.
The second article, “Continuity and
Change: Forms of Liminality in the Sacred
Social Spaces of the Pre-Christian Nordic
World” (Murphy 2016), is a study of the
locations at and into which sacral value was
imbued in the Nordic region during the Late
Iron Age.
1
A range of types of such sites are
identified through toponymic evidence (Brink
2007). Following a discussion of spatial theory
(Foucault 1986; Hubbard & Kitchin 2011),
textual and archaeological evidence for the
social space of an etically-proposed grouping
of sites including rocky places, wetland, and
woodland is examined. The geographic and
cosmological location of these places on the
border of human settlements, and the
subsequent marginalisation of the Other, is
discussed (cf. Hastrup 1990). It is concluded
that the sacral value of such sites was assigned
on the basis of geographic and spatial
liminality. A second grouping consisting of
architectural sites –
hof, cult houses, and halls
– is then considered, and it is noted that such
spaces are cosmologically and socially central
(Brink 1996; 1997; Herschend 2009). It is
argued that the cosmological Other could not
be expected to inhabit such strongly human
spaces, and that their sacral value was instead
imbued on the basis of religious ritual such as
sacral drama (Gunnell 1995; 2011). Finally, it
is suggested that in order for a location to be
invested as sacral space in the pre-Christian
Nordic region it needed to demonstrate
“dimensional liminality”, a sense of
detachment from the human centre of the
cosmos.
201
The third article, “Processes of Religious
Change in Late Iron Age Gotland I: Rereading,
Spatialisation, and Enculturation” (Murphy
2017), is a study of the changes to – and
survivals of – pre-Christian sacral spaces
during the eleventh-century Christianisation of
the Baltic island of Gotland.
2
It departs from
the thirteenth-century
Guta saga description of
how the first churches on the island were
constructed by (the possibly fictional) Botair
of Akubek (Peel 1999). The saga’s justification
for the survival of the second church is
deconstructed, and the Nordic phenomenon of
vé investigated. On the basis of textual and
archaeological evidence (Zachrisson 2014), it
is argued that
vé were spaces charged with both
sacral value and a prohibition of violence, and
featured a range of objective barriers and
subjective boundaries. Gotlandic evidence for
other spaces where violence was forbidden or
taboo is considered, and it is proposed that the
“vi” of
Guta saga was a permanent ritual
location in which Botair deliberately built a
church in an attempt at inculturative
Christianisation (cf. Bintley 2015). Such
Christianisation is argued to have altered the
ontological system within which values were
assigned without altering those values. It is
therefore concluded that Christianisation was
thus a multifaceted process both responsive to
and driven by spatial issues.
The fourth article, “Processes of Religious
Change in Late Iron Age Gotland II:
Centralisation, Enclosure, Privatisation, and
Nationalisation”, builds upon the preceding
work in further deconstructing the process of
Christianisation, and seeks evidence for
notably Gotlandic features of pre-Christian
religion practiced on the island. With reference
to the second article (Murphy 2016), it is
argued that the cliffs under which Botair’s
church stood in
Guta saga may have been the
primary location of sacral value on the site. A
process of centralisation is thus proposed to
have run parallel to the Christianisation of Vi
(cf. Fabech 1994). The strengthening of the
inside/outside binary and the reduction in
experiential access to ritual inherent in the
replacement of an open-air space with a
building is discussed. Evidence from
Guta
saga,
Guta lag, and toponymy is used to
suggest that pre-Christian Gotlandic sacral
places featured a notably public character. This
is suggested to reflect
the notably flatter social
hierarchy of Late Iron Age Gotland (Siltberg
2012; Yrwing 1978). The effects of the
Christianisation are thus argued to have
privatised sacral space to a much greater extent
than had previously been the case, with
concomitant effects on social unit formation. It
is suggested that this forced a renegotiation of
Gotlandic identity. It is therefore concluded
that the Christianisation of Gotland was
accompanied by and achieved via concurrent
processes
of
centralisation,
enclosure,
privatisation, and nationalisation.
The fifth article, “Domestic and Household
Religion in the Pre-Archaic North: Pre-
Christian Private Praxis”,
is an examination of
evidence for small-scale, locally-focused cult
in the Late Iron Age Nordic region. It argues
that such cult can meaningfully be described as
“pre-Archaic” in Robert Bellah’s typological
cultural evolutionary paradigm (Bellah 1964;
2011). A paradigm of pre-Archaic domestic,
familial, and/or household religions is first
established on the basis of comparative studies
of antique Near Eastern and Mediterranean
religion (Bodel & Olyan 2008; Albertz et al.
2014). A household articulation of pre-
Christian Nordic religion is then identified on
the basis of textual accounts from medieval
Iceland. This cult is proposed to have typically
been performed in or near the dwelling; to have
been dedicated more often to localised
supranatural beings (including ancestral
spirits) than to more widely-known deities; to
have offered more significant
roles for women
than other pre-Christian Nordic religion\s; and
to have been more common in the late autumn
and early winter. It is argued that neither food-
based rituals nor the use of iconographic
representations of the supranatural allows the
drawing of useful differentiations from other
pre-Christian Nordic religion\s. The lack of
evidence for
rites de passage explicitly linked
to a household-based congregation or domestic
setting is also noted, as is the possibility that
the picture of pre-Christian Nordic household
cult that emerges in this study represents a
largely west Norse, late pagan articulation of
private religion. It is concluded that this model