My article on abductive inference in archaeology can be found in the most recent issue of Muinaistutkija (3/2013). In it I argued for the fallible and open-ended nature of archaeological epistemology. With some examples from the history of archaeology I attempted to show that the progress of archaeological science is not founded upon inductive data collecting or deductive logic of proving one’s point, but rather that archaeological hypotheses have always been more or less dependant on our abductive powers. On the one hand abductive inference owes to fast and intuitive decision-making, triggered by for example when faced with an unexpected find. On the other hand archaeological generalizations can be characterized as abductive. Because the material facts (or what we tend to take as facts) are often scarce, archaeological generalizations are based on observations from multiple datasets that are not necessarily connected in any causal matter. Comparing two collections and seeing similarities in (not only in the iconic sense of the term) or connections between them is an abductive inference rather than (purely) an inductive one.
All in all, what I tried to say in the article is that even in these times of postmodern multivocality we should not dismiss the role of experts. While it takes time to master even one tiny area of any field of archaeology we should remember that it also takes time to develop bad habits. That’s why the unexpected has such a pervasive role in the epistemology of archaeology.