Let us stipulate, for the purpose of discussion, that open access to the scholarly literature is a Good Thing for the individual scholars and for society as a whole. Why then, do scholars not make their articles available through open access? In particular, why are institutional repositories so poorly populated? (See, for instance, "Institutional Repositories: Evaluating the Reasons for Non-use of Cornell University's Installation of DSpace" .) The question is not idle; as Harvard University contemplates setting up an institutional repository, it behooves us to make sure that the effort is worthwhile and that a significant fraction of the scholarly article output of the faculty end up available therein. I will review the background on the issue and then make a proposal that I believe could lead to extremely high availability rates at modest cost. The proposal does, however, require the enthusiastic participation of the free culture movement.