The purpose of this paper is to critique the empiricist approach to philosophy of science from a scientific realist perspective. Moreover, empiricists' criticisms against realism will be adequately handled by an epistemology that gives priority to explaining how the world is and also by refining the theoretical tools at realists' disposal, such as approximate truth and a causal theory of meaning. I describe an epistemology that provides a basis for realist critiques of empiricism, and then argue that the old realist criticisms can be structured into a stronger view against empiricism, in particular, against the evidential indistinguishability hypothesis. I show how approximate truth makes sense of the history of science for realists. I then look at the strongest arguments for empiricism that come from van Fraassen's constructive empiricism and offer rebuttals to his main points.
