I think the real problem is that you believe that the rules supersedes the DM's right to prioritize the narrative when it feels appropriate.
It's based on the "reasoning" that the DM doesn't like imagining a zombie knocking a hydra prone in his own head.
So, the point of the excercise is players having to come up with an explanation you like of why their character can do what the rules say they can? What's the point of that? Why not just start from the position that their character can do what the rules say they can, and, where you (as a whole group) feel the need to have an explanation of how this happens, work to produce a mutually satisfying explanation collaboratively? Why do you need the GM to be sole arbiter of what is "believable" and what isn't? How does that even make sense?
On this point I tend to agree with Balesir and Dannager rather than NoWayJose - I don't see that the GM has any special entitlement to have his/her suspension of disbelief preserved.
I think that legitimacy (can a snake be knocked prone, or a hydra) is decided on a group by group basis.
<nip>
But in 4e GM fiat in combat has to be transparent. If my power says 'this knocks the target prone' then the GM has to explicitly tell the table that effect hasn't happened. It isn't prone, or it isn't stunned, or it isn't dazed.
I can can understand that transparency not suiting every GM or every group.
This is a good post, and for me at least helps highlight
why 4e lends itself better to a "group consensus" approach to verisimilitude rather than privileging the GM's view. Given that the game is committed to transparency, why not have everyone work together in the way that Balesir describes in the quote above.
The DM might want to say "no" for more reasons than "Because I say so!"
My hack tells the DM to say "no" when the action breaks the consistency of the game world. It's the DM's job to maintain that consistency, because he can - and must - make impartial decisions. Players can't make those decisions because they must advocate for their characters. The DM doesn't have a conflict of interest; the players do.
I'd probably not allow a zombie to knock over a hydra on Lost Soul's internal consistency grounds
<snip>
There is a huge difference between, "We will at all times play with my personal whim," versus, "We will at all times have some internal consistency."
The problem is, what happens when the player decides that an action is consistent but the DM doesn't? The DM over rules the player and the player is basically SOL. The idea that a player is automatically more biased because he's playing a character ignores the large amount of DM's out there for whom the setting is more important than the players. It is quite possible for a player to be just as impartial as the DM.
I think that LostSoul's idea of the GM as arbiter of internal consistency (developed, as I understand it, from some ideas posted by Vincent Baker) is interesting. It's true that, in a typical RPG, players are locked into an "advocacy" role that the GM need not be. But I think Hussar is equally right to note that, in practice, many GMs find themselves advocating also - for their story, for their setting, for
their conception of fictional consistency.
This is why, and particularly for 4e, I think that group consensus makes for a natural solution that reduces the risk of arbitrary GMing, while still placing a check on player advocacy. In practice, for many if not most groups, it is likely that the GM will take the lead in establishing the group consensus (at least, that's so for my group - maybe I'm just projecting onto others in assuming that we're typical!), but there is (in my view) a lot of difference between
GM as lead negotiator and
GM as law.
There should not be "gotchas" based on mechanical rulings. (Situation gotchas are another matters.

)
I think the parentheses in this post are a great summary of what the 4e GM is for (in my view, and as I said a few posts up): providing the conflict that shapes the character of the PCs' "victory". Running the encounters is also a big part of the GM's role, of course, but to some extent perhaps subordinate to designing them. (Of course, if you GM on the fly then designing encounters and running them become more integrated processes. I think skill challenges in 4e can work very well like this - combat encounters can be a bit trickier, depending on how much you as a GM depend upon prepared encounter maps, terrain etc.)
Players dictating the opponents actions in combats is mildly annoying to extremely aggrevating depending on the circumstance and never really felt appropriate even when I was the player and not the GM. It always felt like the combat became about status effects and not the story/narrative reason for the combat in the first place.
This is different than in prior versions of D&D as there was a saving throw for any sort of status effect (with the exeption of a couple overpowered spells) and they weren't available to everyone.
<snip>
Is the GMs role in a combat to interpret player actions and decide how the NPCs respond to the player actions; to challenge the players and provide excitement ? Or is it merely to roll dice for the meatbags until the inevitable conclusion?
If it's the latter than the GM isn't really required for the combat, might as well go read a book for an hour until it's over.
On the other hand, I think this really understates the role of the GM in 4e encounter resolution. Even if we put page 42 to one side - which is a big "if" - the GM has to do a lot more than "roll dice for the meatbags". Deciding what "the meatbags" do, and adjudicating the consequences of PC actions (especially important for skill checks, but often important for powers also), is part of the "story/narrative" and "deciding how the NPCs respond" to PC actions.
I feel is that 4e seems less difficult to run. There is less room, in my opinion, for individual DM talents to shine through. If five 4e GMs were run a module, it'd probably turn out roughly the same... if those five GMs ran another module for another system, I imagine there'd be a larger difference in the play experience.
Nice point - but unless he's running a published scenario with very boring monsters exactly as written, I don't see how that can be so - even in the combat encounters.
For the reasons just given, I tend to agree with Balesir, and think that Wik here understates the role of the 4e GM. But as I said, I think that 4e GMing is most importantly about scenario/situation design.
I've personally never run an adventure module "straight from the can", but would find this a particularly unsatisfying way to run 4e, I think, because it would tend to reduce the game to exploring someone else's story. And in a game in which PC victory is mechanically favoured in the way described above, there wouldn't even be the suspense of failure!
OK, this is NOT going to be politically correct, and I'm probably get a ton of heat over this, but I think it has be asked...
What's at the heart of the problem raised by the OP?
Was 4E designed the way it is partially to appease a new narcissistic 'Me' generation?
4e is not played by a single generation. RPGs are played by all ages. 4e is played by plenty of players who've been playing since OD&D
I'm with Thasmodious here - though I've been playing only since Moldvay Basic rather than OD&D.
I don't see why an approach to the game that favours consensus establishment of the fiction is narcissistic. To me it seems open, friendly and generous!
I just feel the rules are there to serve the game, not the other way around.
And personally I find this phrase slightly annoying. Yes, the rules are there to serve the game. But it doesn't follow from this that the rules are therefore to be ignored - because perhaps following the rules is part of the game! For me, to date at least, this has certainly been the case for 4e. Following the rules has produced an interesting and gripping game. And an important part of that has been applying the rules that provide for player-initiated and player-driven improvisation and narrative.