[The following is a translation of an old post from my Swedish blog]
Abstract philosophical questions is not something that only people with philosophy degrees are occupied with. As a social scientist one must sometimes reflect on some philosophical issues as well. Firstly, it might be good to think about what separates “science” from other human activities, and secondly, about the difference between natural science and social science (and maybe the humanities is also worthy of some attention).
The definition of science is probably not something that normal people lose any sleep over; but active scientists do probably need – at least from time to time – to assess whether a particular study should be regarded as scientific or not. In general, it is probably the case that a certain degree of empiricism has been prevalent since the birth of modern science (in the 16th and 17th centuries). In other words, hypotheses are scientific if they can be confirmed by observation, often in combination with mathematical/logical reasoning. Things that cannot be observed by our senses (e.g., mythological creatures) are not included in the domain of science.
It seems that this way of doing scientific studies has mainly been developed to make controlled experiments the gold standard for scientific rigor. We might, for instance, put some rats in a laboratory, control for as many environmental variables as possible (their food, their sensory impressions, and so on), and then observe what effects we get by systematically changing one of the variables (in cases where we cannot control for the variables we must attempt to settle for historical data, for instance). In the end, what most scientists are interested in is finding certain connections, i.e., what causes what. Do rats feel happy when eating sugar? Can apples cure cancer? Do children who have not been breastfed become violent adults?
This view of science was probably propagated most succinctly by the so-called logical positivists (who started ravaging around the 1920s, influenced by the early philosophy of Wittgenstein, among others). They believed that all statements that are not about empirically verifiable things or about purely logical deductions can be ruled out as unscientific – or as “metaphysical nonsense” if one wants a sharper formulation. The task of scientists should, thus, be to formulate hypotheses that could be verified by observation (which, by the way, has been criticized by Karl Popper, who thought that one should try to falsify hypotheses, instead of verifying them, because even if a hypothesis has been verified many times, it is still the case that one single contradictory case might destroy it.)
The standard view is probably that “pure” logical positivism is not something that all too many scientists (or philosophers) subscribe to, although the science being done from day to day still seems to move in the same precinct. The task is to attempt to show certain causal connections through empirical observation and to verify/falsify the hypotheses regarding these causal connections. Biologists or medical researchers who are interested in other things will probably be the subject of some skepticism among their colleagues.
For the logical positivists there was one, and only one, scientific method. Regardless of what we are studying, – it may be nature or society – we should use this method if we want to be scientific. If this positivist view of what is “scientific” is something that many natural scientists can, more or less, agree to, then one might ask: is it something that social scientists should agree to? Do social scientists think that one should apply the term “scientific” in the same way, regardless of the the object of study, or does “scientific” mean something else when one is engaged in the study of human societies and cultures?