Realism is one of the pivotal concepts in philosophy and is being discussed in various philosophical contexts. Two levels of the realism debate are of particular interest for us: The metaphysical realism debate in analytical philosophy and the debate about scientific realism in philosophy of science.
The metaphysical realism debate poses the question for reality in its full and most fundamental form. Is it possible to give meaning to the concept of an external reality beyond human perception and is such a concept necessary for a consistent philosophical world view? Arguably both sides of the debate are more convincing in their criticism of their respective opponent than in bolstering their own position. Anti-realists argue forcefully that the realist has no clear concept of what he means by external reality. His explanations are based on inadequate and not well defined extrapolations from human intuition, whose ability to provide a comprehensible idea of a world independent of human perception remains a chimera. The realist, on the other side, has good arguments for his assertion that a philosophical position that dismisses any foundation beyond human perception must end in radical relativism (which is actually adopted by French deconstructivism) or radical solipsism.
The history of the metaphysical realism debate abounds with attempts to find an intermediate position between the poles of metaphysical realism and radical anti-realism that is able to avoid the pitfalls of both sides. Grandfather of these efforts is Kant's 'Ding an sich', which does posit a transcendental world but puts it untouchably beyond the reach of human comprehension. Important attempts in younger history are W. V. Quine's naturalised ontology and H. Putnam's internal realism. It is often doubted however, whether any of the known concepts of intermediate positions, if analysed carefully, can be convincingly delimitated from both sides.
While the scientific realism debate is concerned with a narrower question than its metaphysical pendent, its lines of discussion are quite similar. Point of departure is the consensus that the objects of our perceived world are real in some sense, irrespectively of how this might be embedded in the metaphysical debate. The question is whether invisible scientific objects like electrons, quarks or the big bang should be considered real in the same way as their visible cousins. The scientific realist asserts exactly that, while the empiricist understands scientific theories merely as tools to structure and predict the visible phenomena. For the empiricist an assumption of scientific objects thus can be useful at best, but it cannot correspond to any real fact like the existence of electrons in the external world.
Once again both sides of the discussion have strong arguments to refute their respective counter-positions. Scientific realists doubt the possibility to draw a meaningful line between visible and theoretical objects. They also argue that a purely empiricist conception of science is unable to explain its enormous success. Anti-realists on the other side argue that scientific theories are always under-determined by experimental data (i. e. there are many theories which fit each data set), which must render a theory's claim to be true unjustified.
The debate about scientific realism also knows attempts to find a middle path between the two poles of the discussion. Once again however a generally convincing intermediate position has not yet emerged.