Wait a minute. In this case, you yourself are stating that the rules don't work. My assumption (that the +2 social attack stunts are intentional and should be used as examples of how to make social attack stunts moving forward) solves your problem (though it may add new problems, like are social stress stunts still worth double that of social attack stunts, and if not then how do social stress stunts remain competitive). How is siding with broken rules better for play?
Your interpretation doesn't make the rules work, it just makes them broken in a different direction.
The rules are contradictory. No interpretation can reconcile them. Attempting to do so is futile. My problem is insoluble, barring errata or a new edition.
So you have to make the rules for yourself. And my experience suggests that applying normal restrictions to social stunts leads to greater balance and fun. Perhaps your experience suggests differently, but that's beside the point.
Not sure what you mean by "lower level characters" - there are no character levels. Unless you're equating "levels" with refresh spent...and, if you are, experience shows you're incorrect. (In my experience, players tend to either spend all but one or two refresh or only spend one or two. The game tends to disinsentivize a middle ground.)
Chest Deep, Feet In The Water, that's what I mean by levels. Your average werewolf is probably around Up To Your Waist. OW gives a generic werewolf 7 Refresh worth of Powers and 4 skills at Good.
Your average Wizard is Submerged, and has at least one Refinement. (Your Story says each Wizard should take Refinement at least once.)
So when you play a Submerged werewolf, you play an exceptionally skilled example of werewolfkind. When you play a Submerged wizard, you play an ordinary White Council member.
Hmm, I could have phrased that better. Put it this way, in direct conflict outside of an ambush or alpha strike scenario I suspect the caster will win at least seven of ten times. Casters get many of the advantages of an alpha strike / ambush without the need to set up aspects in bunches.
I would agree, actually. Werewolves get better mobility, endurance, and skill selection though, so it more or less balances out. Thaumaturgy's vague level of power makes it hard to judge properly, but choosing a werewolf over a wizard is at least defensible from an optimization perspective.
I'm not saying that werewolves are stronger, but I am saying that they aren't clearly weaker.
Have to disagree here...outside of an ambush scenario where the human can set up a bunch of aspects, a WCV will generally mop the floor with a human.
I don't think so. I'd be up for a playtesting fight.
Not really sure where you're going with this...but it contradicts your stance on WCVs and humans. After all, WCVs get just about everything werewolves do without having to change shape. On top of that they get direct mental manipulation.
The direct mental manipulation is a weakness, mechanically. It costs Refresh, and has nothing to do with your core competencies. Taking both Strength and Incite Emotion isn't really a good idea.
And changing shape is very powerful. Human Form is only mildly annoying (in my experience) and the Refresh bonus is well worth it, while Beast Change is just awesome.