So a lot of the differences that matter in D&D get abstracted away in AW. That's good game design and it works for AW but it's not what I want for D&D. My very favorite part of D&D is to use Player (not PC) wits to figure out a creative solution to a problem and then try to get into some Cunning Plans/Harebrained Shenanigans to make that creative solution work while things descend into chaos.
Then you are inherently asking for rules that are either contradictory or incomplete.
Consistent, Comprehensive, Concrete: Pick two. Want rules that are consistent and comprehensive? You're going to
have to rely on abstraction because otherwise you'll miss way, way too many edge cases. Want rules that are comprehensive and concrete? They're
going to run into problems where the concrete rules tell you to do something dumb, aka, inconsistency (and that's assuming they're actually designed well.) Want rules that are consistent and concrete? You'll be specifically and explicitly choosing to throw the game-runners (DM, GM, ST, whatever) to the wolves whenever they want to do anything outside of those limited rules.
Abstraction is not the enemy. It is an incredibly useful and powerful tool. Like any powerful tool, it must be used with finesse. Unlike the other two concessions--inconsistency and fragmentation--it is not only possible but eminently practical for the at-the-table DM to make an abstraction more contextually concrete based on the situation at hand. That doesn't require amateur game design; it just requires practical human reasoning and sensitivity, something any person can develop simply by interacting with others.
That's just not the emphasis of AW so it scratches a very different itch from D&D for me.
All for all of the "rules don't say" bits you have a choice between either:
1. Abstracting away the differences between the different words so it's more one size fits all (the AW solution).
2. Only allowing a few words (the 5.5 solution). This works well for a more tactical focus.
3. Writing up voluminous rules for how to parse every possible verb (I don't think anyone wants this solution).
4. Have the DM make a bunch of naughty word up within certain parameters.
Some people on this thread have gone over the downsides of #4 in extreme detail. Those downsides exist but I just think that for a game that scratches the itch that I want D&D to scratch for me the downsides of #1-3 (especially 3) outweigh the downsides of #4. Many people on this thread have agreed with me, many have disagreed with me. That's fine. I'd like D&D to be enough of a compromise that there's enough of #4 to make me happy but not so much that other people are overwhelmed.
My assertion is that you have over-emphasized the problems of 1 (which are eminently resolvable without having to constantly act as an armchair game designer) while downplaying or even denying several of the clear and quite obvious problems with 4.
But "the DM will now make naughty word up" has always been an important part of RPGs and always will. There are always different things that the DM will make up and that's what makes RPGs special, you have a human brain making decisions for what happens, not just a set of rules. Now how much scope the DM has to make naughty word up (rightly) varies from game to game, but every game will have it, often in different areas and expressed in different ways.
But what does "the DM will now make crap up"
actually mean?
Does it mean "the DM will completely and wholly reinvent what tasks are, what they mean, how they work, etc., etc.,
every single time such a task comes up?" Does it mean that the DM is now beholden to every precedent and proposal they've ever made in the past, without any record thereof other than player and DM memory, something quite common at most tables?
Or does it mean that the DM is expected to use the tools and concepts presented to them creatively?
Because the former--either capricious (or even malicious) altering of the world beneath the player's feet, or the accidental doing of such because no human can be expected to keep
that much in their head forever with only intermittent reminder--is quite clearly something I think
even you would clearly oppose. But the latter is something everyone, even I, would support.
You can't pretend the two are the same thing though. They aren't. You can have the latter without the former--and the
judicious use of abstraction is one of the most essential tools for making that happen.
Let's check Pathfinder 2e:
Command - Spells - Archives of Nethys: Pathfinder 2nd Edition Database Nope, fails the litmus test. There I just saved myself a whole bunch of time reading review of PF 2e.
Let's check out Shadowdark: "You issue a verbal command to one creature in range who can understand you. The command must be one word, such as "kneel." The target obeys the command for as long as you focus." Hey, this one might be worth checking out.
Saves me a whole lot of time.
And jettisons a vast, vast realm of really excellent game design because you're unwilling to consider the possibility that abstraction is a useful tool, not a burden to be expunged at the earliest possible opportunity.