1script - 5:47 pm on Jan 26, 2013 (gmt 0)
There's very clear logic behind it: to be able to disavow a link, Google needs to keep at least one or two records in databases: one about the original link and another about it's status. I am assuming they are separate because they are created separately and seem to be parts of separate retrieval system but perhaps they are combined - still they need *some* storage space for the link. Multiply that by the number of datacenters and separate index storages and add whatever other info they chose to store for the link and you get a respectable amount of storage space and therefore power, physical space etc. that Google needs to spend on "remembering" a link that's certifiably bad - even the webmaster on the receiving end of PR said so. Then multiply that by the unimaginable gazillions of such links they already store in their system.
In my gut, I feel that Google wants best efforts made to clean up the old links first, and somehow that's going to override everything else. This is just a gut feeling... no strict logic to it.
It costs them absolutely *nothing* to demand that you remove the link and expend *your* time, effort and sometimes money doing it and it costs them *something* to keep the link in the system - as a publicly traded US corporation they are bound by law to demand the former.