Pessimistic Meta Induction.
Scientific antirealists believe that we do not have good enough reasons to believe in the existence of some theoretical entities, that is entities which cannot be verified to exist by direct observation. This is often because of the argument of Pessimistic Meta Induction (PMI). PMI is the argument that because history is full of examples of scientific theories that, despite making a number of empirically verified predictions, were later shown to be false, we should infer from this that our current scientific theories will probably be proven wrong too. Newtonian physics, for example, was certainly successful yet Newton made assumptions that are inconsistent with the theory that succeeded it, Einstein's general relativity. This would mean that Newton's theory is now considered wrong and so we should infer that relativity is probably also wrong.

Realists such as Putnam and Boyd argue that theories are "typically approximately true" because the central terms of past theories refer to the central terms of their successors. Richard Boyd, argues that the truth of scientific theories can be inferred because history does not show successions of false theories, instead "later theories typically build upon the (observational and theoretical) knowledge embodied in previous theories" (Boyd, 1984, pp.41-2). Science is progressive because the terms used in mature sciences refer to the same things as preceding theories, despite having different theoretical definitions.

Poincaré suggests that the structure of theories carry over as well as the empirical content (Poincaré, 1905, pp.162).  Structure refers to the basic mathematical equations used which often appear as limiting cases in succeeding theories.

Brown argues, however, that there can be successor theories which refer but are unsuccessful so reference alone is not sufficient to show that theories are true. It is also likely that reference is not even necessary for truth as many successful theories do not refer to preceding ones. Brown argues that Putnam and Boyd's explanation leaves any theory that is successful without reference a miracle since they claim that success without truth is miraculous.

Cartwright argues that the predictive success of science is an accident that arises from what Berkeley called the 'compensation of errors': "Adjustments are made where literal correctness does not matter very much in order to get the correct effects where we want them; and very often… one distortion is put right by another" (Cartwright, 1983, p. 140). This means that numerous false scientific theories compensate each other to appear true when considered as a whole. Cartwright believes that the laws of physics lie because they do not tell the whole truth. This would, however, still leave novel predictions unexplained.