Yes Dual Wielding incurs massive penalties in D&D. There is a good reason for this: it's because they were trying to accurately reflect real life.
Have you ever tried to fence with more than one weapon? Or fight any fight at all with more than one weapon? It's hard. For exactly the same reason that juggling is hard. The human brain simply isn't naturally wired to simultaneously keep track of two or more things functioning independently of each other at the same time. Most Florentine instructors will tell you flat out that you cannot even start to learn to fight wielding a weapon in each hand until you are equally proficient at fighting with one weapon in either hand.
So do I find those feats unnecessary? Not really, no. It's much easier to swing a baseball bat than it is to duel with two swords.
Well, actually the reason I cited Two-Weapon Fighting first is that I
have tried to fence with two weapons, and it is as the other posters described: the small, off-hand weapon is an opportunity thing. You still mostly use the primary hand, but having something light in the off-hand and using it isn't any more difficult than using just a rapier, at least in my experience. I would compare using two swords to juggling (theoretically, I haven't tried it), but equating using a sword and a dagger to juggling is just silly.
There's also that the penalty for two-weapon fighting without the feat is
minus four for the main hand and
minus eight for the off hand with a light weapon, and the penalty is two points worse with a medium one. I'm sorry, but there is no way to justify that steep of a penalty, even if the PC really was juggling. The -2/-2 that you take with TWF while using a light weapon in the off hand is still a significant penalty, and -4/-4 is even moreso. It is plenty enough to represent the difficulty of it.
So wait, a paladin is a guy who fights with his weapon and casts holy magic at the same time? I thought that was a cleric?
I also thought the whole point of the paladin was a martially inclined crusader-for-justice/knight-in-shining-armor archetype. I'm honestly of the school of thought that paladin's should even have a spell list, just a set of god-granted abilities (like lay on hands and turn undead). Things that can duplicate the effects of spells, sure, but without having to prepare them ahead of time.
The thing is, the reason I suggested Battle Blessing is
because of the crusader archetype. Allowing him to use magic as a swift action means that he is going to actually be fighting while he gets those bonuses, rather than just standing there casting a spell and then going back to fighting afterward. As for lay on hands/turn undead vs prepared casting, thematically I don't see any difference between the two. A god-granted ability is a god-granted ability. Preparation for a paladin is like prayers and oaths sworn at dawn to his god, the abilities his gods grants reflect which oaths he swore and what he asked for in prayer. That's how it feels to me, at least.
The point is, swift action magic lets him do the whole "martially inclined" thing, as he can still fight on his turns even if he uses his divine power. To me, when a paladin is spending a standard action to Turn Undead he is behaving much more like a cleric then when he quickly invokes divine magic in the midst of melee, at which point he is being more like a paladin.
The question then arises; if these feats that shouldn't be feats provide no noteworthy mechanical benefit then why bother making the class features? Why not just remove them as feats (and prerequisites)?
The other problem I have is where to draw the line. You could easily end up with a list of every single printed feat that provides almost no noteworthy mechanical benefit that people don't take simply because there are better options. While that's inherently not a bad thing, if you take them away as feats and instead make them class features it sort of flies in the face of the concept of a modular character design system.
I agree with that part. If you go back to my original post, you'll notice that all of the ones I used as an example did provide a significant mechanical benefit.