Explanatory Unification

Citation: Kitcher, Philip (1981) Explanatory Unification. Philosophy of Science, Vol. 48, No. 4 (RSS)
Internet Archive Scholar (search for fulltext): Explanatory Unification
Tagged:

Summary

The aim of this paper is to develop and suggest the unofficial view, which regards explanation as unification and can avoid some shortcomings of the covering law model. Based on Hempel’s definition, scientific explanation, especially theoretical explanation, is based on objective understanding that achieved by a systematic unification. Kitcher thinks that scientific explanation consists of a proposition and an act to answer questions. Moreover, Kitcher supposes that arguments supplied by science will be used in acts of explanation. Based on his viewpoint, science provides a reserve of explanatory arguments, which we may select when we need. The appropriate arguments relates to a set of accepted sentences. Unification here means that one pattern of argument provided by theory can be used in the derivation of a large number of sentences.

Kitcher proposes the concept of ‘general argument pattern’ to express the structure of the derivations of the same scientific theory by using linguistic terms. Based on his definition, the general argument pattern consists of schematic argument, a set of filling instructions, and a classification for a schematic argument. A schematic sentence here means an expression which obtained by replacing some non-logical expressions in a given sentence with dummy letters. In other words, arguments usually have similar logical structure, in which non-logical vocabulary could be replaced. However, the problem is which explanation we should accept to explain given that there are certain sentences to be true. He thinks that explanation should be the set of arguments which best unifies a given phenomenon.

Kitcher uses systematization’ of K to call any set of arguments that derives some members of K from other members of K. The criterion for best systematization is the notion of unification, which is achieved by generating a large number of acceptable arguments or conclusion about natural phenomena by using a few, convincing patterns. That is, all the patterns may share a common core pattern, which is what we need for explanation. However, Kitcher also recognized asymmetry problem, which means one of the derivations can be used to explain something while the other cannot. He thinks this problem was caused by accidental and irrelevant generalization.

Theoretical and Practical Relevance

I think some ideas are interesting and relevant to archaeology. For example, the structure of general argument pattern Kitcher proposes is similar to the model we use for archaeology. Based on theoretical framework, we develop our model by replacing some non-logical vocabulary. In addition, the idea of unification justifies that archaeology can use theories from other disciplines because of sharing of core pattern.