# Seminal, accessible work on defeasible reasoning

I am taking a graduate-level course in a business-y/sociology-y field, and the logic presented in the literature is typically of the following format "X is a relatively simple and reasonable cause for Y, because reason_1,reason_2,reason_3. Also, we did a couple of case studies and this plays out."

It's not really inductive reasoning, because the case studies are wildly insufficient to alone support the hypotheses of the paper. It's not deductive reasoning because it is not proven that X causes Y.

Coming from a math/science background, I am good with mathematical proof, I am good with a rigorous statistical analysis, but the reasoning employed in this field is foreign and weird to me.

From some poking around, it looks like this type of logic is called "defeasible reasoning". Where can I read up on the epistemology behind defeasible reasoning from a source that would be accessible to a math/science guy?

EDIT heres an example of what I mean. A lot of claims are made, supported by invalid* arguments and a handful of case studies.

*invalid in the technical sense

• Unclear. What are r1, r2, and r3? Also, what does it mean to "science the hypotheses of the paper"? – Bram28 Apr 10 '18 at 17:33
• plato.stanford.edu/entries/reasoning-defeasible follow the bibliography. That being said, this is still definitely a case of inductive reasoning, contrary to your second sentence, it's just not as strong of an example of induction as is used in, for example, physics or chemistry. – Not_Here Apr 10 '18 at 19:27
• @Not_Here, I would disagree about the inductive reasoning bit. The reasoning does not follow "We witnessed X cause y in some circumstances, therefore X causes y generally", the meat of the reasoning is defeasible, and the case study is only used to do a sort of post hoc sanity check on the defeasible bits, imho. – Scott Apr 10 '18 at 19:39
• This is called abductive reasoning now, a.k.a. inference to the best explanation, see Are “if smoke then fire” arguments deductive or inductive? that links to some references. The old terminology did not distinguish between this and the more conventional induction, and some are still using "induction" broadly, both are of course defeasible. – Conifold Apr 10 '18 at 19:44