Archaeologia Islandica - 01.01.2007, Síða 91

A REASSESSMENT OF THE DEVELOPMENT OF THE CATHEDRAL AT GaRÐAR, GREENLAND
side. Furthermore Norlund believed that
the nave of the church was expanded in
this phase (compare fig. 4 and 5). The lat-
ter interpretation is, however, in my belief
open to question, particularly regarding
the nature of the south wall of the nave.
Here Norlund would ascribe the parts of
the south wall which are not coloured in
figure 4 to Garðar 2. Consequently the
south wall of the nave became markedly
skewed in comparison with the earlier
church and it would have meant that the
nave was almost 2 metres wider at the
eastern end than at the western end (Nor-
lund 1930, 32) (exterior measurements).
Furthermore it would have skewed the
chancel itself towards the north when
seen from inside the church, creating a
definitely unbalanced church interior that
carmot have gone unnoticed.
While it is not impossible that
Norlund’s theory was right, it does beg
the question of why the builders were
satisfied with this crookedness of the
church. Certainly it seems unlikely that
the skewed south wall was a result of
inept builders, since no other Norse
Greenlandic stone church is as markedly
crooked as Norlund pictured Garðar 2.
It seems even stranger when keeping in
mind that the church at Garðar was the
cathedral, and not just a regular parish
church. I would suggest that the founda-
tion which Norlund pictured as belonging
to Garðar 2 could instead be traces of a
supporting wall, erected at a later point
to ward off instabilities in the south wall
of the nave. This would mean that there
was no major increase of the width of
the nave in phase 3, but that the church
retained the width from phase 1. Phase 3
is thus represented solely by the further
extension of the chancel and the erection
of the side chapels.
Here another point must be made.
As mentioned earlier, Norlund believed
that there was a door in the corner
between the nave and the south chapel.
This was based on his belief that wall A
(fig. 6) belonged solely to Garðar 1, and
was not used as a foundation when the
extended chancel was built (cp. fig. 5). It
is also contingent upon the fact that the
nave was indeed widened, since the small
opening that Norlund saw as a door was
formed partly by the foundations that he
ascribed to the widened nave. Even if the
nave was widened, which I doubt it was,
I have already stated why I do not believe
in the existence of a door at this place. I
would further suggest that the part of wall
A, which lay inside the south chapel, was
utilized in connection with the extended
chancel. Now the earlier foundation for
the enclosure wall was used as a founda-
tion for the west gable of the south chap-
el. This would cause the south chapel to
mirror the north chapel closely, which
seems more likely than Norlund’s propo-
sition. Summing up, I would suggest that
phase 3 consisted solely of the extension
of the chancel: it was lengthened and two
identical chapels were added, one at each
side.
Phase 4
In connection with phase 4,1 will return
to the east gable of the nave, which was
discussed in connection with phase 1.
As mentioned there, Norlund considered
the middle part of this foundation to be a
later addition, inserted in connection with
Garðar 2. It is never clear in the publica-
tion why Norlund believed this to be the
case, and it is not altogether impossible
that the nave had a foundation running
the entire length of the east gable from
the outset. If the middle section of this
89