If I understand Karl Popper's reasoning correctly, a theory is 'scientific' if there can be devised a test that tries to disprove it; i.e., it must be falsifiable.

Now, imagine a theory that theoretically we could build a test to disprove it, but it is not possible to actually build such a test, e.g.:

  1. It requires some significant computational power that is not possible to assemble.
  2. It requires some technology that is not known to exist today ('build a black hole and check')
  3. It is immoral to build such a test environment

Would such a theory still be considered 'scientific'?

4 Answers 4


According to the way things are defined, a theory is 'scientific' or not—not the test. So if one can devise a test that can falsify a prediction of the theory, then the theory is scientific.

If it is impossible right now to create such a practical test then, no, the theory is not scientific. If you can't actually perform any kind of test (for whatever reason), there's no way to get falsifying information, and so the theory cannot be falsified and so is not 'scientific'.

But a theory that has impractical tests is much more in the direction of science than something without even the possibility of falsifiability with unlimited resources. It eventually may allow falsification. Also, one's falsification test may have its own problems (one has a subsidiary theory about how the test works), and it may take engineering of that subsidiary theory which will then enable it to falsify the main theory.


I think that is kind of an open question. The criticism of string theory is that it cannot be tested in an practical way. So critics say on those grounds, it is "not even wrong." It fails to be a scientific theory.

  • As far as I understand it, ST could feasibly be tested at sufficiently high energy levels (the string-like nature of fundamental particles would become measurable) -- we just do not have the tools to get to those levels yet.
    – Joseph Weissman
    Oct 13, 2011 at 23:04
  • Joe, if it cannot be tested at all then it doesn't fit the assumptions of the OP, the theory must have the possibility of being falsified, just current circumstances don't allow a test to take place. @Joseph: My take on string theory is that there is a lack of testable hypotheses, that is it is just mathematical creations/derivations. Under your analysis, it fits the OP's assumptions. Do you then have an idea of whether ST is then scientific?
    – Mitch
    Oct 14, 2011 at 2:13
  • @Mitch I am certainly not an expert so I hesitate to weigh in much further than my comment. I might suggest that it is scientific in many ways, in terms of the rigor of the formulation of many of the conjectures. It is my understanding that at high energy levels the string-like nature of fundamental particles would become increasingly clear -- there would be new specific resonances, etc -- but I just do not think we have the equipment to generate the requisite energy (at least for the time being.) I would love to hear from an expert on this point.
    – Joseph Weissman
    Oct 14, 2011 at 2:38

I think a better way to frame the question is "Is a theory that is not falsifiable (in practice) useful?"

Put this way, I think the answer becomes more clear. If we adopt Jamesian/pragmatic terminology, we can say that the "cash value" of a scientific hypothesis resides in its testability (and the reproducibility of the results of that test); without those, it remains merely a thought experiment.


I think the answer is yes, because the question whether a theory is scientific or not should not be dependent on time and resource constraints. If a theory is not scientific at some point of time, it can't suddenly become scientific once the human race acquires some resource. Morality is also something that changes with time and culture.

  • There could be another level of 'scientific' 'not scientific' and 'quassi scientific'
    – yarony
    Oct 13, 2011 at 20:01
  • @yarony: I think the word you're looking for is Pseudoscience :P
    – stoicfury
    Oct 14, 2011 at 4:15

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .