> Have you read this particular version of the saga in ON?
> This was the first time that I really pursued an ON text of this length,

It was the first long text that I tackled too. I used
Turville-Petre's edition published by the Viking Society [
http://www.viking-society.group.shef.ac.uk/publications.htm ], which
has a very helpful glossary and notes. He says that the text he uses
"has been prepared and carefully normalized by Dr. Guðni Jónsson"; as
far as I know it's essentially the same as the version you're using.
Of course, I still made lots of mistakes on my first attempt, many of
which could have been avoided by paying closer attention to that glossary!

> and I found that I had some inaccurate expectations. To some extent,
of course, this is a result of the author not always using perfect
grammar.

I think the author was in a better position to judge what was
'correct' Old Norse grammar than we are, well than I am anyway ;)

> Some of the inconsistency between the precise linguistic
construction in the language lessons and the text thjat I observed
appear to be the function of fairly prevalent conventions. For
instance, singular present third person form of vera, er, is
frequently employed at the beginning of sentences in disagreement with
the tense and number of the rest of the sentence, so that directly
translated would appear "the berserkers is went onto the shore".

Hmm, I can't find a sentence that matches this exactly, but maybe
you're thinking of 'Ok er þeir broeðr koma í Sámsey, sjá þeir, hvar
tvau skip liggja í höfn þeiri, er Munarvágr hét' "And WHEN those
brothers come to Sámsey, they see [where] two ships [are] lying in the
harbour [that was] called Munarvágr." (I would leave out the words in
square brackets if I wanted it to sound more natural in English.)
Besides "is", 'er' can mean "when" (= þá er), or "which".

> The use of relfexive/middle voice forms of verbs in this text
really did not correspond very strongly to their usage described in
the language tutorials they had become familiar to me in. I don't know
if this is unique to the text or a commonality. It was often used in a
manner that showed essentially no distinction from an active form of a
verb.

This is perfectly normal. The suffix has all sorts of idiomatic uses
with different verbs. Sometimes it simply adds a reflexive or
reciprocal sense, but with some verbs it can correspond to an active
verb in English. As Blanc Uoden mentioned, a lot of these verbs
describe a situation where the agent is also the patient, whether the
agent does something that affects themselves (klæðask "get dressed"),
or the subject is more than one person doing something to each other
(eigask við "to fight"), or sometimes the suffix is used for mental
events: fælask "to lose heart, to panic"; minnask "remember" (cf.
minna "remind").

> I really love this passage, and spent a fair amount of time on it.
Its structure makes it very difficult to translate; I considered "then
light shone from it such as from a ray of sun", "then it shone with
such light as from a ray of sun" even though light isn't mentioned.
"then shone from such as from a sunbeam" seems like the perfectly
literal translation.

Great isn't it! The bit where the berserks arrive at the island is
another favourite of mine. The more wordy U version often loses the
fantastic momentum of R. I agree an extremely literal translation of
this would sound wrong in English because of the impersonal verb. As
you suggest, I added the word "light": a light shone from it like a
sunbeam. A simpler alternative that's just occured to me: "it shone
like a sunbeam". I think this might be better, since a ray of sun IS
light, and it sounds a bit convoluted to talk about light shining FROM
a ray of sun.

> This is generally the case with all translation, but there is
always some degree of a subjective element in terms of to what extent
text should be rendered literally and to what extent it should be
construed as a composition that sounds natural in the language in
which it is being translated.

I guess a lot depends on the purpose of the translation, whether for
illustrating grammar for people learning Old Norse, or making
something that recreates as far as possible the artistic effects of
the original (including metrical effects)--which raises the question
of what ARE the artistic effects of the original--or with some other
aim in mind. For the first function, I like the way modern textbooks
and papers on grammar often print a literal translation followed by a
more natural and idiomatic one: the best of two worlds, at least.
Since this group is concentrating on learning the language, we tend
not to discuss aesthetic details so much, or haven't been doing
recently, but it's an interesting topic.

Llama Nom