Yes I agree that at the mining level or forking the block chain with enough computational power is an issue (inherent with any coin using a similar consensus model for verification of transactions), BUT that is not the issue we are discussing as that is merely determining the block/ledger for which a coin stores transactions. The item of topic was "privacy" and how Dash's approach has been an "attacker economist" approach.
As far as I am aware, the "attacker economist" approach would also be used for dealing with the possibility of Sybil attacks in a cryptonote coin.
The fee structure would also be an "attacker economist" approach to prevent the blockchain from being bloated to DOA levels (and the coin / transaction network going to the grave), due to the problematic scaling and bloat of ring signatures at high mixing level.