Even if it were a characteristic of all undead, 4e runs on exception based design, your unique undead wouldn't be breaking a rule, it would simply be an exception.
Lol. This is hilarious when combined with:
If a DM wants to change the rules and say swarms can't be grabbed, but they can subjected to involuntary movement, that's a house rule, it might influence some players choices or it might not, but it's up front, the players know what it means.
On the one hand, you're advocating "exception based design" and then on the other, you're saying it's a "house rule". One in the same, imo.
Ok. Fine.
But, let's get one thing super-straight.
I never said "all swarms cannot be grabbed" - please, go back and read over my posts. That's all I ask. Read carefully.
In sum, I said, "In some cases, a swarm may or may not be able to be grabbed. It depends on the fiction. DMs, if you believe a swarm should not be grabbed, write 'Immune: grab' in its stat block. If a player comes up with a maneuver that breaks a rule, like no forced movement, use DMG page 42 to let them try - if it makes sense in the fiction (like taking a board and pushing back a medium swarm of rats). To do something mechanically, do something fictionally. And, if they do something fictionally not represented by mechanics on their sheet, don't disallow it - use page 42 of the DMG."
There's no "arbitrary" method behind this. It's what's happening at the table, at the moment, and what is going on in the fiction. If a rule needs to be broken in order for that to happen, so be it. If the rule swarms cannot be forced to move by a melee needs to be broken because it makes sense in the fiction - I will break it. It's not "on a whim" because the immediate happenings at the table matter. It's not arbitrary because it's directly tied to fictional circumstances and the rules as written as a guideline.