Has anyone tried to classify what is considered to be unknowable in the physical world to define it more precisely? I can think of some basic classes of ideas and some, perhaps disputable, examples for each class.

  • Lack of information transfer: What happened before the big bang? What happened on a specific day 100,000 years ago if no historical data/information exists about that day?
  • Paradoxical/Self-referential: One cannot experience their own death; one cannot see one's own eyes without a reflective surface etc.
  • Senses: Experiencing a spatial dimension beyond 3. Anything beyond the five senses.
  • Mathematical: Godel's Incompleteness Theorem and other properties of arithmetic formal systems.

The question linked below mentions Fitch's Paradox. I'm afraid I don't have the background or understand of notation to comprehend the article at SEP. If someone could briefly summarize and explain its relevance to my question I would appreciate it.


Is there a philosophy of things that could not be discovered?

  • 1
    What happened before the Big Bang is not necessarily unknowable, unexperiencable can be knowable too, most of modern physics is beyond the 5 senses, and undecidable sentences become decidable in stronger theories (e.g. Godel sentence of arithmetic is provable in set theory). Isn't it a paradox to know of something unknowable? As Wittgenstein wrote, "a nothing will serve just as well as something about which nothing can be said". – Conifold Sep 16 '16 at 18:37
  • Something may be beyond the 5 senses but in the case of science, it is because it there is some kind of information available for us to infer. Generally, when I hear about interpretations of before the Big Bang I seem to see a lot of "we can't know because there is no way to test it". I don't see it as a paradox to know of something unknowable... there is a difference to me between asking/having knowledge of an unknowable question and getting the answer to that question. Even if my categories are wrong, what then is unknowable? – syntonicC Sep 16 '16 at 19:03
  • "We can't know" because we can't test it today. We can not know today what we can or can not know tomorrow, there is no known "unknowable". On realist philosophies there can in principle be unknowable things (like Kant's thing in itself), but by their very definition there is nothing we can know of them. On other views (e.g. Peirce's pragmatism or phenomenology), "unknowable" is an unintelligible term. I think what you have in mind are horizonal things, we get hints that they are there but do not have means of grasping them yet. This is a relative and time specific notion, "known unknowns". – Conifold Sep 16 '16 at 20:26
  • See Traub's interview:"What are some of the reasons why a scientific question might be unanswerable? I'll limit myself to just three here. The first is that insufficient data has survived. That can be a problem in ur-linguistics, archaeology, and history. The second is that contingent events, sometimes called frozen accidents, may limit our ability to explain certain phenomena... Finally, resources, such as energy, may simply not be available in our part of the universe to discriminate among contesting theories about the universe". – Conifold Sep 17 '16 at 21:59
  • Regarding "unknowable" being an unintelligible term. I think I understand: We can't call something unknowable because the only way to know if it was unknowable or not is to know about it. But if it can be known then it is was never "unknowable" to begin with. Traub's position seems to bear some similarity to my original list (insufficient data/information) but I guess he isn't saying its unknowable, just not answerable, which I guess I considered to be the same thing in my list. – syntonicC Sep 20 '16 at 14:48

Kant arguably classified things in this manner, although not with the same categories as you suggest. He has two categories: Things as they appear to us through the mediation of our sense, or in other words, empirical reality, and noumena, things as they actually are in an unknowable objective reality. The idea is that all we can know is what our senses tell us, but that we can only assume that what our senses have told us is accurate.

Fitch's paradox is an attempt at a logical proof of the following statement:

If there is at least one unknown truth, then there is at least one unknowable truth

The main idea is that we can't know "X is an unknown truth" because that requires us to know "X is a truth", which equates to knowing X, which means X cannot be unknown. Therefore "X is an unknown truth" is unknowable, yet true (in the case that any such X exists). This is yet another category, things that are logically unknowable.


Kant used the word "noumenon", which has not been accepted by all dictionaries. The word can refer to that, which was previously "unknowable", being "thing-in-itself", thus, that, which is underneath phenomenal shapes (forms), which are known. As science progresses, phenomena are becoming more, and noumena are becoming less, because newly discovered phenomena replace parts of previously unknown noumena.

  • I voted this down for being a misunderstanding of Kant. – user20253 Sep 12 '17 at 9:48
  • Would you support your opinion with an excerpt from Kant self about noumenon. Preferably not a paraphrase. I support my opinion with Kant's own description "thing-in-itself". Modern science delves deeper into insides, therefore logically, new phenomena replaces noumena. You probably have the more metaphysical cosmological view of noumena, which in my view is not in the spirit of Kant's philosophy. He demarcated metaphysical views away from science and regarded philosophy as a scientific ontological process. Added to that, I am not a specialist on Kant. Read only two books of himself. @PeterJ – Marquard Dirk Pienaar Sep 12 '17 at 13:17
  • I recommend this essay. It is not quite fair to Kant imho, but it is an excellent overview.and much easier to understand than anything Kant wrote. philosophynow.org/issues/118/… – user20253 Sep 12 '17 at 16:56
  • @PeterJ. Thank you, but I am interested at this time of my learning to read only what Kant wrote himself. – Marquard Dirk Pienaar Sep 12 '17 at 17:57
  • Fair enough. I'd say it's the hard way to get to know him, but also the most scholarly and respectful way so good on you. The idea that 'newly discovered phenomena replace previously unknown noumena' is, however, utterly wrong. – user20253 Sep 13 '17 at 12:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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