Has nothing to do with bad DM's and everything to do with the mechanics. It's not that older games weren't fun most of the time, it's that older games forced a very traditional setup of "act on your turn, not on someone else's".
We've suddenly got real time combat?
I don't have the books in front of me, but the 'Attack of Oppurtunity' concept which forms the core of the argument you are advancing here is not a new concept. It has just been formalized better than in older editions, but in older editions taking a non-combat action while threatened did draw an attack from anyone threatening you. In fact, by the time 2e came out, I'd taken this notion and formalized it into 'parries' and 'counterattacks' that looked alot like a clunky version of 3e's AoO. But even without my formalization, it's in there. You couldn't make a missile attack or quafe a potion in melee without getting attacked.
Moreover, older game systems actually in some ways did a much better job of handling simultaneous action than modern versions of D&D. Most elegantly, older versions of D&D maintained the wargame-like concept of 'phases' within the turn, so that for example, everyone made simultaneous movement during a movement phase, which was followed by everyone taking turns making attacks. In practice, this actually made the battle simulate real-time much better than 3e and post 3e's strictly turn based sequence where everyone completes a full set of actions and attacks in initiative order. So I'm not sure I agree that there has been net movement in the direction of acting out of turn, except possibly in 4e's concept of 'team play' where your actions can trigger actions in your teammates. That is new, although I'm not sure that it necessarily means you act more often since if you aren't the one who gets an additional action, it just means you are waiting longer for your turn. On the whole, I tend to think of it as a wash.
Older, traditional games spent a HUGE amount of time doing not much of anything. You sat and watched the other players doing something a whole pile. You couldn't act out of turn.
This seems like a false assertion backed up by a thin bit of evidence. The primary time in older traditional games I can remember spending a huge amount of time doing not much of anything is when we split the party - and that was only for some groups.* Most groups played a more traditional game and never split the party, and in that case I don't have the recollection you have. Splitting the party in a modern game will end up with the same passive participation phase as you watch (or not) the other members of play do their thing while waiting for yours.
*(Splitting the party in some groups became a big enough problem that in some groups I had a primary PC and several 'hench-PCs' who were the henchmen of another primary PC. This resolved what had been the increasingly big problem of waiting around watching other players do something.)
Heck, going way back, you had a caller who would mediate between you and the GM, removing even your ability to really decide all your own actions.
I go back far enough to remember when this was the suggested course of action in the rule books, but not far back enough to remember this ever being strictly enforced. And, even in the example of play in the 1e DMG, it wasn't strictly enforced and there is direct DM to non-leader player communication. So, I think that you are speaking more from theory than from actual play experience here.
It might be hyperbole, but, there is a grain of truth there. Older games, by delaying gratification, meant that you were spending more time not being gratified. This should be pretty obvious. If the rate of gratification is higher now, then it must have been slower before, with more time spent between points of gratification.
There is a grain of truth here, but I think it is more in the occurance of certain sorts of failure and the expectation of failure than in the rate of play. High level 1e play is typically much faster than high level 3e play, and I would suspect at least as fast or faster than high level 4e play because the system did not encourage so much attack modifier management, nor did you typically have as much to do on your turn so rounds typically went faster. More to do in a round vs. getting to your turn more quickly seems like a wash to me.
Thus the hyperbole of 20 minutes of fun wrapped in 4 hours. You spend a minority amount of time actively participating, and a majority of time passively watching and waiting for your turn.
I disagree. There were times when you passively participated because you were diseased, unconscious, stunned, or had been tied up because the psionic blast had driven you insane and you kept trying to gut yourself with your own sword (yes, I am speaking from experience), but that wasn't 'a majority of the time'. I don't think you are going to make much headway on the participation argument via the standard model of combat.
One thing that more recent games have done is break out of the turn based systems that characterize many traditional games. Many games now have some sort of interrupt ability (to borrow a CCG term) that lets you "go out of turn". Many games have abilities which allow you to make someone else take an action out of turn as well. Whether it's something as simple as an Attack of Opportunity mechanic or something more complex, the idea of "wait your turn" is a good example of delayed gratification.
I suppose so, but I'm not convinced there is as much progression here or that the 'interrupts' you talk about are as significant to what has changed as you seem to want to make them.
Since gratification doesn't equal succeeding, but rather actively participating, again, I have to ask, is this trend really a bad thing? Forcing more active participation from the players more often is a good thing isn't it?
I don't agree with the premise, and to the extent I agree with the conclusion it wouldn't mean that I was also validating the premise because the one doesn't to me imply the other. Actively participating in failure is less gratifying than actively participating in success (for many or most players) and even ungratifying regardless of the level of mechanical participation so I don't agree in your attempt to split the concepts. And I don't agree that there is a increased participation trend except where failure is defined down. Since I don't agree with the premises, what I think about encouraging involvement in the game doesn't much directly impact what is being discussed here.
Moreover, I think based on the argument you made here that you don't even get the whole premise of the claim "20 minutes of fun in 4 hours of play".