Random fixation and/or founder effect, among a myriad of effects that don't involve natural selection.

>Let's say for instance that a species of ant is known to be more aggressive, why it was that way is probably generic drift. Why it stays that way is because of its environment, and it is selected for.

Or the different alleles for aggressiveness (whatever that means, and assuming they exist) got fixed due to drift? Who knows. The burden of proof is on the adaptationist to show that something other than contingency happened.