Sometimes in science you just have to use your imagination.
In 1905 Albert Einstein published a paper about how atomic particles move through fluids and what happens to molecules when heat is applied. It was the invention of the term “atom” and “molecule.” It was a pure hypothesis that (at the time) couldn’t ever be tested, and indeed, would never be actually observed for over 100 years. Einstein’s untestable theory would win him the Nobel Prize (NOT Relativity).
In 2013, the advances in electron microscopy allowed us for the first time to see at the atomic level, albeit very fuzzy. The picture below is from the U.C. Berkeley electron microscope, that shows a Rubicene (Hydrocarbon) molecule of C26H14 being held together by a Benzene “ring” (C6H6). It’s used in photosensors and solar panels. It essentially says that Einstein was EXACTLY right in his 1905 paper. The full Berkeley findings can be found here.
Out of Einstein’s untested hypothesis of something that could never be directly observed or tested came the invention of Propane, Butane, and atomic energy, among other advances. Oppenheimer didn’t observe a Uranium-235 atom shedding neutrons, rather he was able to measure its radioactive effects of the things around the atom, and just assumed that it was what Einstein had predicted.
In other words, in natural physical sciences, hypotheses aren’t always tested because they are not always testable. Instead theoretical science tells us that if the theory is correct, then this is what we should look for.
In the Post-Positivism argument, social science either tries to be like physics or runs from it as fast as it can, under the idea that natural sciences always hypothesis-tests things. This is not how natural sciences work. Yes, hypothesis testing is good, but as Einstein showed us, it isn’t everything. Sometimes you just have to use your imagination. Beyond the imagination, as Thomas Kuhn pointed out, sometimes there is no such thing as an “objective” observer. Sometimes we displace Einstein’s atoms as we move through the fluid to observe them.
Instead of the label of “post positivism” in describing the social sciences that run away from the natural science, perhaps the better label is anti-positivism (PDF). Post-positivism does not reject the scientific method, but rather modifies it to include subjectivity. Anti-Positivism completely rejects science. Durkheim, like Einstein started with observable facts, and asked questions from there. Anti-positivism, or “post-modernism” completely rejects the idea of observable facts entirely.
Yet facts are important. It really is true that Benzene rings tie hydrocarbons together. It really is true that African American males are the lowest paid social group in America. It really is true that the sun revolves around the Earth. The gender wage gap really exists. These observations are the same: natural facts versus social facts. No amount of anti-positivist or post-modern theorizing is going to negotiate that the facts really do exist.
The problem for natural sciences versus sub-atomic social science isn’t the math, or the methodology. I think it’s more basic than that. Math is used as a TOOL in the natural sciences to explain HOW things work. Math is used in positivist social sciences to explain WHY things work the way that they do. This is problematic. In one, math is a tool. In the other, math is the end result. The more basic issue is the question of why versus how. Theoretical science can address both, but often doesn’t in the social science.
Just as Einstein told us what to look for in his theories, theoretical social science rarely tells us what to look for. Instead, they tell us “why” we observe the things that they do, without explaining how. For example, in Einstein’s General Theory of Relativity, he gave us within a very definite measure, of what to look for in the universe – the bending of light by space. He said that IF his theory was correct (and it was), then light should present itself in a very specified way. Yet in social science, we observe, and then assign “blame” by assigning a theory to the phenomena.
In contrast, social science theory doesn’t tell us what to look for, or explain the “how” of the phenomena. For example, absolute systems of male power is to blame for the misogyny. Yet, it fails to explain those (growing number of) males who refuse to exercise that power. The natural cycles of the market are to blame for recessions. Yet markets are man-made, and manipulated. Discrimination is to blame for high minority unemployment. Yet African American women have twice as much chance of getting a job than African American men.
That doesn’t mean that these phenomena do not exist – they most certainly do. It just means that the theory isn’t really telling us what to look for, since we already find observable anomalies outside of the theory. That also doesn’t mean that developing a “grand theory” is the answer. It does let us know however, that the way social science “does” theory is a little broken. Social science is trying to explain “why” without explaining “how.” How can theory tell us “why” without explaining “how” large anomalies exists that defy the theories? How can a theory tell us “why” without telling us what to look for?
Whether social science goes toward positivism or away from it, either direction will result in social science being more like natural science, especially where the imagination is used. Social science needs to be okay with the idea that some ideas are not testable – maybe even for a century. More to the point, social science needs to become okay with the idea of not knowing. Where social science cannot go if it still wants to be considered a science, is in the anti-positive (or anti-science) direction, otherwise it will no longer be a science, but rather an ideology. We’ve had a lot of social problems created by ideologies masking themselves as “science” already.