I'm afraid I know nothing about philosophy. However, for a part of my course I'm currently studying the mind body problem. Many of the theories I come across can neither be proven nor disproven by scientific means. Is there a word to describe this phenomenon?

  • 2
    Conjecture ? ...
    – Mauro ALLEGRANZA
    Commented Dec 6, 2015 at 19:07
  • 2
    @CamilStaps. Given the context of the question, I don't think satisfactory definitions would necessarily be found at an English Language site. There may not be a satisfactory "word," yet a helpful reply. This seems like the right place to ask.
    – Nelson Alexander
    Commented Dec 6, 2015 at 19:32
  • 2
    If it can be neither proven nor dis-proven then it is non-scientific.
    – Hot Licks
    Commented Dec 7, 2015 at 1:14
  • 3
    It is unfortunate that this is too old to migrate because again this is a question looking for a domain specific technical language. Philosophy.SE is the best place for this.
    – Mitch
    Commented Jun 8, 2018 at 13:32
  • 2
    @Mitch That’s where it came from. I don’t know whether the migration to EL&U can be rejected after all these years. Anyway, it looks just as acceptable as any other SWR here.
    – Lawrence
    Commented Jun 8, 2018 at 15:31

6 Answers 6


the term you are looking for is unfalsifiable, according to Collins Dictionary:

unable to be shown as false, although possibly not true

Attribution: Collins English Dictionary. Copyright © HarperCollins Publishers


First, I am not aware of some "technical" term for such theories... and perhaps one is not needed beyond "untestable theory," which is, of course, commonly used in science.

In physics, for example, there are hypotheses, such as quantum entanglement, string theory, or parallel universes that arise out of logical extensions of given theories and problems, but present serious challenges for the experimental verification scientists require. In such cases, conceiving of such a possible experiment can require great genius and imagination.

I may have this a bit wrong, but in the case of entanglement, first thought to be untestable, Bell's theorem provided an experimental scenario, recently applied. In the case of string theory, the verdict is out, and in the case of parallel universes, it may be logically impossible to test. Caveat: Again, I am not up on the latest here.

Generally, it is considered "reputable science" to propose new theories only with accompanying proposals as to how they might be confirmed or falsified, as Einstein famously did with general relativity and his proposal for detecting the gravitational bending of light. Because such an experiment had not even been thought of before, and was considered highly unlikely, the positive results found by Sir Eddington carried great conformational authority for Einstein's theory.

Aside from "untestable theory," what else might such explanatory speculations be called? Well...speculations, for one. Or they might be called logically-linguistically "meaningless" if they are proposals with no correlating "real content." They might be dismissed as "metaphysical," if they entail entities without causal or other "physical" properties, and are therefore obviously untestable.

We might also refer here to Kant's "Antinomies of Pure Reason," conflicting propositions that are not susceptible to solution by use of reason, let alone experimental proof. Kant's examples included the existence of God or a "necessary being" and whether or not the universe has a beginning in time. Oddly enough, the discovery of cosmic "background radiation" by researchers is now considered evidence for a datable "beginning" of the universe. So the expanding framework of science may alter what sorts of theories can or cannot be tested in accord with scientific method.

Again, short answer: "untestable theory..." or, for greater emphasis, "bucketful of moonbeams."


Empirical means "derived from or guided by experience or experiment" or "provable or verifiable by experience or experiment" (here).

You might call a theory not testable empirically a nonempirical theory.

Even better, in contemporary philosophical parlance, metaphysical usually connotes being beyond the orbit of scientific, empirical testing. Thus you might call a theory not testable empirically a metaphysical theory.


There is no natural native word in English to describe this situation. In informal thought, it is somewhat vague what 'scientific' means, and 'proven' and 'disproven' sound like math terms.

But in the domain of academic philosophy, these terms have more technical meanings. 'Scientific' has come to mean, over hundreds of years of philosophical argument, using sensory data (and by extension tools that transform sense data) to support a statement.

Also, a universal statement is difficult to prove (you'd have to test everything which is just not possible), but it could be disproven (a single counterexample/a failed test). An existential statement can be verified by giving a single tested example. The first is falsifiable and the second verifiable.

You're wondering what to call a statement that is neither falsifiable nor verifiable. This is at the heart of the 'demarcation' problem between science and non-science (or what is sometimes called metaphysics).

Things which cannot be proven/verified or disproven/falsified are then called metaphysics or metaphysical.

Pejoratively they might be called 'unscientific', 'philosophy'

'Untestable' is less pejorative and leans towards the one-sided 'unverifiable'.

  • How about "indeterminate"?
    – tautophile
    Commented Jun 8, 2018 at 16:03
  • @tautophile You may want to give that as your own answer with extra explanation of why you think that works.
    – Mitch
    Commented Jun 8, 2018 at 17:32
  • Mitch, my response was not so much an answer as a suggestion. Since, in the OP's question, you can't prove or disprove--determine the validity of--a certain theory by scientific means, it would seem that the theory is "indeterminate" or "indeterminable" in scientific terms. However, I realize that using the word that way might be stretching the existing meaning(s) of "indeterminate" beyond acceptable limits.
    – tautophile
    Commented Jun 15, 2018 at 16:02
  • @tautophile I think that word is a good suggestion, but since it is somewhat distinct from mine, I thought it would work better as a separate answer.
    – Mitch
    Commented Jun 15, 2018 at 16:32

All of the following are untestable: all opinion, all normative statements or values, and all metaphysics/philosophy.

Just because things are outside the domain of science does not mean that values, philosophy, normative statements and opinion are not real and/or not important. One might not approve of all such things, but no one can function without any philosophy, normative statements, or values.



From Merriam-Webster:

unable to be proved : not provable

Literally, the word suggests exactly the notion that something cannot be proved (one way or another).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.