The original whale accounts weren't even that big anyway 1-2% (mathematically impossible for them to be much larger when Steemit started with 80%).

Now of course it is possible people have bought more and accumulated much larger (but broken up) stakes by now but I've seen no evidence of it (and not even 1% is needed for the kinds of issues you describe with the social layer).

Maybe it's more a perception than a reality in that those who do have larger stake and use it to excessively self-vote or downvote people they don't like seems more outrageous

Yes you are referring to the dynamics of how the social layer works. I agree not that well. But I don't think you can pin this on the distribution being excessively concentrated. By general economic standards it doesn't appear to be at all. Rather I'd argue the contrary: the social layer isn't capable of handling even normal and moderate degrees of concentration without causing results that many find disappointing. This is a problem with the social layer.

When I last looked at breakdowns of minnows, dolphins, orcas, and whales, it sure seemed like the whales owned most of the currency, but maybe I'm misinformed and that was just being skewed by Steemit, Inc accounts.

Probably skewed to some extent (bear in mind that even excluding the steemit account itself, most accounting of 'whales' include steemit team/founder accounts, part of the original 80% ninja-mine), but look, if you have people with 1-2% each, not really obscene amounts, and there are even 10-20 of them, that's going to be a high % of the total. Just how the math works out. This is basically normal. What percentage of the world's population owns 50% of the wealth? How many shareholders of the typical company does it take to reach 50%, etc.?

Expecting ownership of any asset to not be concentrated at all is very unrealistic.