Structural Paradigm

Observational and Theoretical Statements

In the 1960s, Rudolf Carnap, in his Philosophical Foundations of Physics, published in 1966, found “the distinction between what may be called . . . empirical laws and theoretical laws . . . one of the most important distinctions between two types of laws in science” (p. 256). In his definition, empirical laws refer to “properties directly perceived by the senses” such as “blue,” “hard,” and “hot” from the point of view of the philosopher, or properties “that can be measured in a relatively simple way” from the point of view of practising scientists in their disciplines (p. 225). But what the senses deliver is not precise. For example, “hardness” is not a binary category opposing hard and soft. Rather, the senses allow for gradual distinctions, sensing one thing “harder” than another thing and even sensing that Thing A is harder than Thing B and this in turn is harder than Thing C. Such an experience can similarly be made with concepts like color or temperature, attitude (uttered in everyday language), or even gender (as most languages have words like “effeminate” making clear that sex and gender are not simply binary). Moreover, a property measured in a “relatively” simple way is not really clear-cut, and whether a measurement is “simple” depends on the experience and the disciplinary background of the researcher. For example, a physician needs to have a thermometer at hand to measure a temperature, or a political scientist needs a questionnaire to measure a political attitude such as party identification. To measure these concepts, one already has used some theory about the linear function between a temperature and the expansion of mercury or about the complicated connection between the answers to questions in a survey and the behavior when casting a ballot, respectively. Perhaps the laws mentioned in these two examples are “observational” and not “theoretical” laws, as the linear function assumed in a clinical thermometer is just arbitrary before the underlying mechanism that makes mercury expand is clarified. The same applies to the relation between the communication between interviewer and interviewee and the ballot finally cast by the interviewee (or the answer given when asked for the prospective ballot). In the case of the political scientist, the assumption that there is some function between the two “measurements” is even more arbitrary when the underlying mechanism that makes this interviewee—and many other interviewees—give the two answers is not yet analyzed in more depth.

The problem persists that theoretical laws were formulated in theoretical terms, whereas empirical laws were formulated in observable terms. Moreover, it is not always clear (as in the two examples) whether assumptions such as “the hotter the longer” or “mercury expands by 18 × 105 per Kelvin” are observational or theoretical laws. The former seems to be observational, as “hotter” and “longer” are accessible to the senses; whereas the latter rests on the assumption that the expansion coefficient is equal for all temperatures at which mercury is liquid and thus seems to be a theoretical law. The law determining that interviewees who say that they like a candidate best among a selection of candidates are frequently those who say they will vote for a particular candidate is an observational law; whereas a law stating that there is a positive correlation between the value on a sympathy scale given to a candidate and the propensity to vote for that candidate is likely to be a theoretical law, mainly because “propensity” is not available to the senses.

...