ABSTRACT
Observation plays a unique role in philosophical accounts of the scientific enterprise. Traditionally it is what distinguishes science from other epistemic enterprises like mathematics, philosophy, theology, and many of the pseudo-sciences. Conventional wisdom has it that the content of our observations is given to us by nature itself – it constitutes our data. Hence its pronouncements are mandatory. We may adopt opinions that go beyond what has been observed. But (according to conventional wisdom) these opinions are minimally required to square with the data. Everyone agrees that science goes beyond the observational given to some extent (or else it would be mere journalism or natural history). But different groups of scientists and different historical eras have held vastly different opinions about how far beyond the data it is permissible or desirable to travel. The more closely a scientist or a philosopher hews to the data, the more of an empiricist she is. A major peak of empiricism came in the 1920s and 1930s with the logical positivists. According to the early (and more extremely empiricist) proponents of that philosophical school, a statement is meaningless unless it can be translated, or reduced, to observation language – a language consisting of terms that describe only observable properties of observable things (Ayer 1936). Statements about unobservable electrons were thought to be reducible to statements about observable tracks on photographic plates; statements about unobservable mental states were to be translated into statements about observable behavior; and so on. These translation exercises failed, in psychology as well as in physics. Almost all of the interesting and fruitful concepts of science resisted reduction to anything remotely like an observation language. Their failure impelled the positivists to liberalize their criterion of meaningfulness. The old requirement was that scientific hypotheses must be logically equivalent to an observation statement. The new requirement was that hypotheses need only entail one or more observation statements. Scientific theories were now permitted to contain unreduced theoretical terms, so long as the theories had observational consequences. This became the standard view of science at mid-century. The standard view also encompassed the following account of how one should choose between competing theories of the same domain. To choose between theories T
1 and T2, you find an observation statement O such that O is an observational
consequence of T 1 , and not-O, the negation of O, is an observational consequence of
T2. Then you observe whether O or not-O. The theory with the right observational consequence wins.
