On epistemic possibility: A reply to Hirschkind and Tambar