Observational Advantages: A Philosophical Discussion

摘要

I distinguish two kinds of observational advantages: (i) a given representation is observationally advantageous over another if a logical consequence of the information represented in it is observable in the former but only inferable from the latter; (ii) a given representation is observationally advantageous over another if a logical equivalence is observable in the former but only inferable from the latter. The paper also discusses the following question: observing (vs inferring) a piece of information in a given representation is an advantage if the purpose of the system of representation is to directly observe what could otherwise be inferred. But if the purpose were to infer what could be otherwise be observed, then one should conversely speak of observational disadvantages.