Sociology, philosophy, music performance, almost any pursuit can be a science, if the scientific method is applied. The scientific method is the process of rational-empiricism. You need both sides of that, the analytic and synthetic, theory and practice, logic and experiment. If you begin with a falsifiable question, explore it, and explain it, and test your explanation, then the activity is scientific.
(This is based on my reply to “What is Philosophy? a Status Seeking Answer" by Fabio Rojas on the OrgTheory blog here.)
I find a dose of humility in Tom Lehrer's singing about "Sociology" (... they can snow all of their clients/ by calling it - heh - 'science' / when in fact it's only sociology...), but in truth any pursuit can be a science without mimicking physics. And of course, we all know that physics is taken as the standard. The problem with that is that human beings are not billiard balls. History and philosophy and the all the rest are, indeed, sciences, but you can have two or more very different answers to the same question and have both be valid. “Do the American people support President Obama’s policy in Afghanistan?” is a complicated question, and perhaps a set of questions, but with statistical - rather than absolute - answers.
The scientific method is the process of rational-empiricism. You need both sides of that, the analytic and synthetic, theory and practice, logic and experiment. You can have a useful tool that you do not understand - the steam engine for it's first 150 years; maybe even computer chips today... - but it is not science until you can explain it.
With history, in particular, we cannot experiment on the past, but any explanatory theory can be validated and falsified by testing it against different experiences, e.g., the Greeks and the Mayans; or different records from the same time and place.
My reply is that we understand the unreality of formal logic - "Socrates is an elephant. All elephants are blue. Therefore, Socrates is blue." - but that is trivial and demeaning when contrasted with, say, Andrew Wiles's proof of Fermat's Conjecture for which all the experimental evidence never was and never would never be sufficient.On whether philosophy is a science, Prof. Rojas suggested: "... but falsifiability through logic is qualitatively different than falsifiability through experiment or observation."
Also, seemingly ethereal mathematics has proved worldly: irrational, negative, and imaginary numbers are easy examples. Indeed, the very concept of "number" was an idea with no "reality" as long as our languages differentiated a brace of pheasants from a pair of shoes. In Japanese, we still count sticks (pencils) differently than leaves (pages). The problem with linguistic analysis from Wittgenstein on is that such discussions seldom get past English and rarely exceed the bounds of Indo-European. So, when philosophers attempt to argue “meaning” or “the meaning of meaning” they often stray from what is truly empirical and therefore cannot come to a falsifiable claim. That is not science. But the errors and omissions of some do not invalidate the entire field.
Prof. Rojas also wrote: “In the end, through, I approve of McGinn’s status seeking exercise. Systematic investigation of logical arguments is different than art history or music performance.”Again, the proof of the pudding is in the tasting. I could claim that photography informed painting in the 19th century; and I could argue the contrary as well; but eventually, I must show some examples, i.e., provide some evidence. You don't need to play the piano to figure out how to write a piece that no one could perform. However, you might enjoy reading it; and perhaps, it could be programmed via computer and synthesizer. Again, the performance validates or falsifies the claim. Even music performance can be a science.
While philosophers worried about how we know the sun will rise tomorrow just because it always has, they never worried about “the problem of deduction.” From valid axioms you can “prove anything” even if you never find one -- again, the square root of minus 1, negative numbers, etc. In fact, the development of double entry bookkeeping and alternating current electricity provided the empirical evidence for those rationalist conjectures. But the point is that academic philosophers were more troubled by everyday experience than they were by higher mathematics. The basic problem with academic philosophy - as academic sociology or academic physics - is that it avoids testing.
Objectivist philosopher David Harriman offered a solution to “the problem of induction” in his book, The Logical Leap: Induction in Physics. The presentation left some aspects unanswered. Also, it might have been more productive for Harriman to call his program "objectivism" and leave the word "induction" where it resides, rather that attempting to drag it into his narrative space. His work remains significan and should be considered by anyone who questions whether philosophy can be a science.
Reality is real. If something is logically true - not just a blue Socratic elephant - or perceptually evident - not just a trompe l'oeil - then the logical analytic rational side must support the empirical synthetic experiential side, or the “structure” just will not “stand” and the claim must be regarded as unproved. When philosophy is carried out according to the scientific method, then it is a science.