N.B. The email exchange below may have been edited, e.g. to remove content not essential to the main point(s) or to standardize English spelling/grammar.
Dear Dr. Todd Krause:
Thank you for the helpful answers to the questions in my previous email about your excellent site Old Norse Online.
I do have three more questions I hope you could help me solve, because I was unable to find an answer in the Old Norse Online pronunciation guide.
Thank you again for the wonderful site Old Norse Online. I have been using the site regularly to learn and practice Old Norse, therefore, I am looking forward to your response in order to help me fine tune my pronunciation.
Sincerely, E. G.
Howdy Mr. G.,
My sincere apologies for the delayed reply. Thank you very much for your patience.
In response to your questions, I personally don't have many hard-and-fast answers. But I can give you some suggestions based on inference.
1. From a historical perspective, when we encounter the earliest uses of an alphabetic system, we assume that the scribes write what they mean. That is, if they write 'g', then they mean that sound. If they write 'gg', then they mean that sound twice. If the sound was a stop, then two occurrences mean that stop in succession; if a fricative, then the fricative in succession.
Of course this is all mitigated by considerations of the origin of the alphabet. For example, the Gothic alphabet derives from the Greek, and the latter already had a long-standing convention of writing 'gg' for the sequence /ng/. Gothic therefore adopts the same practice.
Specifically regarding your question of whether gg is a doubled stop or doubled fricative, I don't personally know of any scholarly work that makes a well-reasoned claim one way or the other. The fact that the modern pronunciation shows g sounding as /y/ before front vowels suggests to me that g may not have represented a true stop anyway, but rather a sort of affricate (similar to Modern Greek, incidentally). So the actual pronunciation might be somewhere in between the two alternatives you propose.
2. As for l, if there's any voicing to be had in the positions you mention, it was evidently weak. We can see that in the fact that we find devoicing of final consonants following l: -ld > -lt.
3. In the earliest (perhaps prehistoric) stages of what became Old Norse, it's quite likely that what was later written as v represented /w/. On the one hand, Eng. water is cognate with ON vatn; moreover both are cognate with Hitt. wātar. So they all go back to PIE */w/, which alternates with PIE */u/ between consonants. This suggests that PIE */w/ had no fricative value, and this most likely survived into the earliest stages of Proto-Germanic, and later survived in English. The fricative nature of the ON phoneme seems to be a later development.
So ultimately the question is how early v became fricative in ON. I'm not sure off the top of my head what the answer to that question is. But a quick look at what the First Grammarian writes seems to suggest that in his era (c. 12th cent.), v was still likely /w/.
I hope that helps some with your questions. Best of luck, and as always, please keep me apprised of your progress.