Observations Presuppose Theory

First, observations presuppose some principle of selection. We cannot go on observing anything we come across. We need “relevant” observations. In science it is the problem that decides what is relevant and thus provides the principle of selection. Hence, there cannot be observations without a prior problem. To quote Karl Popper, ‘Before we can collect data, our interest in data of a certain kind must be aroused; the problem always comes first' . It may be objected that the problem itself is due to the observations we make, and hence, observations come first. But, this objection does not hold. Two persons might make similar observations though only one might come out with a problem. This shows that mere observations will not generate a problem. How, then, are scientific problems generated? It is usually when there is a clash between what we observe and what we expect. Of the two persons making similar observations one may come out with a problem whereas the other may not because the former has expectations which conflict with the observations s/he makes, whereas the latter does not have any expectations. The expectations are generated due to our belief in a theory. Thus, problem generation presupposes a prior theoretical commitment. In other words, a prior belief in a theory is necessary for a problem to be generated and a prior awareness of the problem is necessary for making relevant observations.

Secondly, in science, observations are taken into account only if they are desirable in a language that is currently used in a particular science. An observation, which howsoever genuine, cannot be expressed in the recognized idiom for all scientific purposes is no observation at all. It is the theory which provides the language or the idiom to be used in describing observations. It is tempting to quote in this connection words of the physicist and philosopher Pierre Duhem: