One thing that makes this question difficult is that some grammatical distinctions that were originally originally clear and important can get watered down in time and can become vaguely synonymous later on, and we can't be exactly sure which exact stage we're on in each instance when we're talking about a dead language.
I just don't think reified rules ought to be pressed too far
If the rules don't fit, they have to be changed. I mean that there must be some constraint as to when you can use the aorist and the imperfect seemingly interchangeably and when not. Homer isn't ungrammatical.
With the help of Chicago Homer, I went through all the instances of ὤρνυτο, ὤρνυτ', ὤρνυντο, which I list here. This made me refine my ideas a bit.
IL.3.13, IL.10.483, IL.21.20, OD.22.308, OD.24.184, IL.5.13 (?), IL.16.635, IL.19.363. These have durative, or gradual or other easily understandable use of the imperfect and don't concern us for now.
IL.7.20, OD.2.2, OD.3.405, OD.4.307, OD.8.2, IL.23.131, OD.2.397, OD.24.496, IL.3.267, IL.3.349, IL.16.479, IL.17.45, IL.23.488, IL.23.664, IL.23.689, IL.23.759, IL.5.17. These are the difficult ones, where the described action seems punctual. Chantraine (p. 192) calls them "verbes exprimant le développement d'un mouvement" (verbs expressing the development of a movement) , but admits this is "très proche de l'aoriste" (very close to the aorist). Probably these don't all come from the same mold, so there isn't probably one single explanation for all of these. Often it looks like as if there's some emphasis on the result
of the action, rather than the action itself, that something that follows is being prepared. When the character gets up from bed, this "sets the scene" like Qimmik says. When this describes a character attacking another, the description of the inflicted wound follows (Il 5.17). In Il. 3.267, a description of the sacrifice follows.
Here are the first 15 examples of aorist ὦρτο (I only looked these, as they are too numerous):
IL.5.590, IL.7.162, IL.7.163, IL.7.211, IL.8.135, IL.8.409, IL.10.523, IL.11.129, IL.11.151, IL.11.343, IL.11.645, L.12.377, IL.13.62, IL.15.124, IL.15.312.
In all these cases there seems to me to be some emphasis on the action itself, it's immediacy. For example in Il. 7162-163 the heroes seem to be springing up in response to Nestor's speech, not in preparation of something that follows.
I think in most cases here "punctual" imperfect and aorist should be translated the same. What I'm suggesting is that in this case Homer has two vaguely synonymous forms, and while in many/most cases both forms are possible, the poet has a choice; probably there's a pattern to found, if even it's only a statistical probability rather than an absolute rule.
As a Finn, I'm sure you're glad that Finland wasn't incorporated into the Soviet Union in 1917-8, if for no other reason than that you were spared learning Russian numerals and verbs of motion (which have three verbal aspects). But I don't think you're really Finnish--your English is too good.
Thanks for the compliment. If you want to know my linguistic background, Finnish is
my mother('s) tongue, I'm also half French from my father's side. I started learning English at about 11; like in many small countries with strange languages, many people are ok in English here. My spoken English is very poor though, you wouldn't believe. French I speak better than English but read and write less well. Unfortunately I don't know Russian.