> en hann gaf hinum vopn sín (why reflexive?) ... og bað þá fara til

fósturjarða sinna.

In the second example, 'sinna' refers to the unstated subject of the
infinitive clause = 'þá', the direct object in the main clause. But
in the first example, 'sín' must refer to 'hinum', the indirect
(dative) object. There are some similar examples in Cleasby / Vigfússon:

http://lexicon.ff.cuni.cz/png/oi_cleasbyvigfusson/b0529.png

I don't know why this happens, but I've met quite a few examples, so I
don't think it can be all that unusual for the reflexive possessive
adjective 'sinn' to refer to something other than the subject,
although Faarlund calls it "rare and contrary to the general rule".
It's quite normal for 'sinn' and the reflexive pronouns 'sik' and
'sér' to refer to oblique experiencers; examples from Faarlund's Old
Norse Syntax:

þótti honum nú sitt líf meðallagi gott
his life now seemed reasonably good to him

ef honum þykkir sér þat nökkut fullting
if he thinks that (will be of) any help to him

This is one of the reasons why some linguists classify these as dative
subjects. But Faarlund also has one curious example where 'sér'
refers to an "objectlike dative".

ok keypti þar pell ágætlig, er hann ætlaði konungi til tignarklæði sér
and bought there some very fine material which he procured for the
king for his robes of state

I can't remember meeting any other examples like this last one though,
so I think it must be much rarer than instances of 'sinn' referring
back to non-subjects. Faarlund doesn't have any examples of 'sik'
refering back to an indisputable non-subject, so I'm not sure if this
is possible. Unfortunately, he doesn't make a distinction between the
possessive adjective and the pronouns here, so it's not clear if he
his comments about the one apply equally to the other.

LN