Arguments are needed to provide theories but I'm only used to deductive reasoning in physics and science, maybe due to their empirical nature. Does anyone know if other species of arguments have or can be used at a professional level?

These include: inductive, Abductive, Argument by Analogy and Reductio ad Absurdum.


closed as primarily opinion-based by Kyle Kanos, stafusa, Aaron Stevens, Jon Custer, WillO Sep 6 at 14:47

Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ I would just point out that in science, there is not really deductive reasoning in the strict sense of the word. Rather, there is abductive reasoning. We find the most plausible explanations--there is almost never a case where one can unambiguously deduce a model given a set of observations. $\endgroup$ – Dvij Mankad Sep 6 at 10:43
  • $\begingroup$ @Qmechanic Are "philosophy of physics" questions welcome on PSE, say, as soft-questions as long as they are not highly opinion based? If not, I would be glad to delete my answer and vote to migrate the question to Philosophy SE and post my answer there. $\endgroup$ – Dvij Mankad Sep 6 at 11:17

In physics, one doesn't formally recognize or make use of the categorizations of the kinds of reasoning. In fact, to echo the sentiments expressed by Steven Weinberg in "Dreams of the Final Theory, Chapter VII", I would say that it is the job of physics to actually find out as to what kinds of reasoning are conducive to discovering the truth and not the other way around, i.e., we cannot presuppose what would yield the truth. However, one can try to descriptively analyze what kinds of reasoning have been employed in physics so far.

The crude basis of the scientific method can be expressed as follows. $1.$ Noting the empirical data, $2.$ Trying to find the simplest model that can explain the observed data, $3.$ Making new predictions, and, finally, $4. $ The validity of the exercise comes from testing (at least a subset of) these predictions against the experiments.

The second step resembles deductive reasoning but it is not really deductive reasoning because there is never a mathematical way to deduce a unique model from a given set of observations. So, model building is really more like abductive reasoning. Making new predictions certainly involves both deductive reasoning and inductive reasoning. For example, the mathematical steps involved include deductive reasoning whereas the act of making physical predictions about the real world is inductive. There is also some element of reasoning by analogy in the form of symmetry arguments that go into the theoretical process as well as in the act of expecting that if the same experiment is repeated then it would produce the same result. However, this is an oversimplified version of the scientific process. Of course, the theoretical process often involves dealing with multiple models which contradict each other in extreme cases where they overlap, i.e., they create paradoxes. The recognition/formulation of a paradox often involves a reductio ad absurdum argument.

It seems far fetched to suggest that the whole of the scientific method is only some combination of different modes of reasoning. Rather, it seems more in the spirit of the scientific method to expect that the scientific method will teach us valid modes of reasoning.


Reductio ad absurdum is sometimes used in physics.

For example, the orbital theory of electrons. It was thought that perhaps electrons orbit around an atomic nucleus. But since it is known that every accelerating charge must always emit radiation, orbiting electrons must emit radiation and lose energy. So they must continually fall into closer orbits where they emit more radiation until they enter the nucleus. Therefore electrons cannot have stable orbits. They must move only in straight lines at constant velocities. They must be stationary except when exterior forces accelerate them, in which case they emit radiation. Because the only possible alternative -- orbiting electrons -- is absurd.

Similarly, since we know that light travels at the same constant velocity for everyone -- including the source of the light and any observer that interacts with the light -- there is a contradiction. When there is relative velocity between source and observer, the lightspeed must have two different velocities. Therefore it is necessary to give up euclidean space and linear time. Because the classical assumptions contradicted themselves, and this is the only assumption we can give up that can fix the problem.

On the other hand, the reason there are many interpretations of quantum mechanics is that physicists became unwilling to assert that any of them are absurd.


Not the answer you're looking for? Browse other questions tagged or ask your own question.