This was just something I had thought about a few weeks ago and I was just reminded of it while browsing threads and I thought I'd see what you all have to think about it:
In 4th edition, is alignment is largely a tool for NPCs and monsters and not players?
The first thing that made me think this was the fact that there is very little alignment-based functionality in 4th Edition D&D. Classes have been alignment generalized so that you can have lawful good barbarians and chaotic evil paladins if you want to, magic circle has become bounded to creature type rather than alignment, and (as far as I can recall) there are no powers or items that are relevant to alignment itself. Of course, DMs are always free to add their own material that's alignment specific, like saying that certain creature-typed powers also work on evil creatures or that a healing spell from a good-aligned divine healer won't work on a chaotic evil entity, but these are player-created-content rather than part of the system. (The difference can be measured in the number of hours it takes to incorporate the alignment-sensitive system into the game, its effectiveness, and the players' willingness to learn the new rules and incorporate them.)
Deities are divided along alignment lines, but these alignments are largely prescribed with no basis in morality (D&D religions have always been more about philosophies of life or purposeful goals rather than moral exactitudes anyway). If you want proof, read an article on the "white man's burden" before worshiping Erathis.
Exacerbating this situation is the fact that the 4th edition PC side is somewhat more neutral+ in focus than previous editions, which is not to say that evil and chaotic evil alignments aren't accessible to players or that skilled players can't successfully incorporate such characters into games (even good-aligned ones), but as a general rule neutral+ characters tend to be the norm. The Player's Handbook and the Dungeon Master's Guide both engender this covertly (by omitting the evil deities from the PH) or overtly (recommending that DMs run games where characters are neutral, good, or lawful good). The cutting of most good and lawful good monster manual entries is another big step in this direction (though, as a DM, this is a decision that I was thankful for since, as the creators said when they were discussing their philosophy of design, they wanted to have a Monster Manual full of creatures the players would likely be endeavoring to defeat rather than wasting space on those which might occasionally get in the way but were otherwise benign).
Which brought me to my second thought: Is that such a bad thing?
Alignment has always been a subjective entity, which is what it should be: There are no cut-and-dry alignments in real life. It's a sliding scale based on perspective, circumstance, and a few hard-to-nail-down unversal moralities (i.e. a sentient creature has the right to live without his or her own will being impinged on by another sentient creature), which are further complicated by the diverse range of sentient life in the D&D universe. Is a demon truly chaotic evil if it is born to that alignment and has no possibility of ever being anything else? What is the difference between a cat or dog that chases down smaller animals for fun and a demon that murders commoners because it is its nature to do so? Is a hard neutral character evil if he chooses to not act against evil out of apathy or self-interest? Is a murderer who was violently abused throughout his childhood a villain or a victim or both?
Certain entities (the afforementioned demons and angels of deities of the various alignments) have always represented the extremes of the alignment chart (or line, as it is now), but the borders between neutrality and extremity have always been up to the players and DM to decide. (Order of the Stick has a funny bit on this subject with regard to stick-in-the-mud paladins and their more moderate counterparts.) By hanging mechanical gameplay elements on subjective functions
My final question was: If they aren't for players, what function do alignments have for DMs?
My conclusion was that it seems that alignment in 4th edition exists less as a moral stance and more as a shortcut predictor for behavior for creatures. Even this behavioral shortcut can easily be accomplished through the flavor text (and, more often than not, it is required anyway due to the many ways in which good and evil can manifest in behavior and social order). More to the point, intelligent creatures of any species are (generally) capable of being of any alignment, with the only exceptions being those which are expressly incapable of being any other. Even regardless of their base stat block alignment, if a creative DM wants two creatures to be buddies (or enemies) he could probably find a way to make it happen. Common goals, shared enemies, or even a passing interest in one-another might lead two alignment-incompatable creatures to work together towards a single purpose.
So even this conclusion is tenuous. Really, there doesn't seem to be any purpose for alignment anymore other than as a means of dividing religous beliefs along morally ambiguous lines. Flavor text is more functional in terms of defining NPC and monster behavior and it is considerably more multifaceted and exacting to boot. Players tend to be better at defining alignment through actions rather than labels, and the flexibility of the alignment lines tends to make such labels all but useless anyway.
I'd love to read what others have to say on the subject, but at the moment, my final conclusion is: Absolutely nothing.
Maybe it's not such a big deal worth making a point about. Then again, maybe we would have gotten a few more stat blocks in the Monster Manual if we'd dropped alignment altogether, or perhaps had more dynamic, fleshed-out creature motivations for various species (distinguishing orcs from gnolls or kobolds from goblins in more significant ways than the manner in which they rampage or the interchangable higher powers they revere).
Sorry if this bit at the end got less refined: I'm sleepy and I may not be representing my argument as well as I was when I started. Oh well.