Full Idea
To defend type-type identity against the multiple realisability objection, we might say that a molluscs's brain events that register pain ARE of the same type as humans, given that being 'of the same type' is a fairly flexible concept.
Gist of Idea
Maybe a mollusc's brain events for pain ARE of the same type (broadly) as a human's
Source
PG (Db (ideas) [2031])
A Reaction
But this reduces 'of the same type' to such vagueness that it may become vacuous. You would be left with token-token identity, where the mental event is just identical to some brain event, with its 'type' being irrelevant.