Obviously, sometimes D&D is about combat. 4E is the big one here. If you're not spending a decent chunk of your sessions fighting monsters in 4E, you're playing waaaaay outside of what its designers intended for it. This is a function of the complexity of the rules 4E has related to combat: in 4E D&D, you have a lot of interesting decisions to make inside of combat, and you're not risking much by initiating it. Your guy has a lot of neat things he can do in combat, the rest of the party has a lot of neat things they can do in combat, and the monsters, terrain rules, encounter design guidelines, and other DM advice and features of the game make it fairly easy to set up a particular kind of "interesting encounter." And the DM has to step outside of what the game rather strongly recommends to even be able to kill the PCs. It's tough to do accidentally.
Depending on the quality of the GM, of course, you probably have some interesting decisions to make outside of combat as well, but it's harder to say exactly what you're risking in those situations than it is when you're working within the game's combat system. 4E D&D is a game where you're supposed to spend a lot of your time hitting things.
At the opposite extreme, you've got OD&D/Basic and their retroclones. Combat in OD&D isn't that interesting unless the DM knows what he's doing and the players are active and creative-- neither the games rules nor its advice really funnel play towards "fun" combat. To the degree that combat is interesting, it's interesting because you're allowed to bring in whatever non-combat systems you have for handling problems-- the nets, 10-foot-poles, and spells of physical problem solving-- into the combat. If that's fun in the rest of the game, you're probably going to have fun with OD&D combat, too.
The main function of the combat rules, instead, is to make combat deadly, in a way that's fairly adjudication agnostic. If the DM is doing her job right, she's going to kill your character sometimes, and you're going to know that you deserved it. It needs fairly detailed combat rules because it's relatively difficult to adjudicate combat compared to most of what you do in D&D, and relatively important compared to most of what you do in D&D that it be adjudicated "correctly," or at least in a fairly neutral way. (Among other reasons-- to a degree combat is always complex because combat is inherently interesting).
So here complex combat creates a situation that's the opposite of what it does in 4E: "If we get into combat we will probably die; as long as we stay out of combat we might die but we're not sure" vs. "If we get into combat we probably won't die; as long as we stay out of combat we might die but we're not sure." It doesn't change the fundamental D&D situation of "you don't really know what the DM is up to, and to play and have fun you have to be willing to trust her."
The combat rules create a particular environment for decisions to be made, and they create context for decisions made outside of combat. In 4E, combat is inherently interesting, and the context it creates encourages players to engage in combat. In OD&D, combat isn't inherently interesting, since its intended to create a context that discourages players from engaging in combat without pissing them off if that's what they decide to do. In each case you get different behavior. (Depending on the assumptions that the players themselves bring to the table. I've had players fling themselves into OD&D combat because they didn't understand the rules and they assumed that it was like the video games they were used to, or other games that they'd played.)
For me this is a simple and clear case of some general principles in game design. Just because a game has a lot of rules for something, it doesn't mean the game wants you to spend a lot of time doing that something. Player's may assume that's the case, if they mistake "complex" for "interesting," but they'll eventually learn better if it's not. If a game has a lot of rules for something that's a good sign that it's important, but it may be inherently important or it may be important because it's a failure state or other consequence of normal game play. 3E has fairly complicated rules for death and dying. That doesn't mean that 3E is "about" death and dying to the degree that it has rules about them.
As an aside, I feel like this combat/rules dynamic puts 3E in sort of a weird place. The volume and kind of rules that it has for combat indicate that combat is inherently interesting. For the most part, that's true. But it lacks a lot of the safeguards that 4E has built-in to combat. It's not nearly as deadly as OD&D, but it can still be pretty deadly, especially when the players misjudge the situation somehow. It extends a lot of OD&D's assumptions to their logical conclusions-- OD&D combat can be interesting, if the DM presents it in a sufficiently textured way, and the players have some toys to play with, so D&D 3E provides the texture and the toys.
Unfortunately, that makes it easy for things to guy awry if the players then take their inaccurate OD&D assumptions to their logical conclusions. Either they'll die a lot and get frustrated, or the DM will low-ball the challenge (and the 3E books don't give a whole lot of guidance away from this tendency) and the combats will get really easy. The underlying physics of D&D's combat can make for pretty boring combat if there aren't any interesting stakes involved; if you're not trying to achieve something in particular, or desperately avoid death. 3E makes it interesting to go "oh hey! this game is about combat!" then have that initial impression confirmed, and play that way until the game gets very boring and the DM gets annoyed.
