> But now I'm all confused over the idea of "idiom"

Here's one definition, from the Oxford English Dictionary: "3. a. A
form of expression, grammatical construction, phrase, etc., peculiar
to a language; a peculiarity of phraseology approved by the usage of a
language, and often having a signification other than its grammatical
or logical one."

Not sure whether that helps; it's not really an absolute term, at
least not when used in this way...

> How often do words have such different senses when they are pleural
> compaired to singular? I tend to categorize those types of shifts as
> idomatic... but if it's really commmon/regular I guess it wouldn't
> be...

I suppose these could be called idiomatic, since they depart from the
norm shared by English and Old Norse for the meaning of the plural to
be predictable from the singular, and do so in ways that are usually
pecular to Old Norse. Or you could sidestep the whole 'idiom'
question and just regard them as pairs of separate words, one of which
just happened to have historically been the plural of the other. A
couple more that come to mind:

lag, neuter sg. layer, fellowship, tune, etc.
lög, pl. law, laws.

lát, neuter sg. loss, death.
lát, pl. manners.

Similarly we have a few words in English where the meanings of what
were once plural and singuar of the same word have diverged, e.g.
'mean' : 'means'; 'air' : 'airs' (as in "to give oneself airs"); and
for some speakers 'medium' : 'media'.

There are also quite a few 'pluralia tantum' in Old Norse, words which
only occur as grammatical plurals, some corresponding to singular
nouns in English, e.g. 'jól', neuter plural "yule". Obviously
language is a quirky thing, but it's only when we start to compare
languages that we realise just how quirky!

Hope that hasn't added too much to the confusion...