Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Proof proves nothing

  1. Jan 28, 2004 #1
    Anyone familiar with Willaim James' critique of "vicious intellectualism?"

    James says that concepts are static, incomplete abstractions that are at best only useful analogies of dynamic reality. The corrolary is that reality cannot be completely described or captured by concepts.

    "Proof" therefore is essentially nothing save a series of concepts that explains reality to your satisfaction.

    But in the final analysis, what you have is not the guaranteed truth, but only the satisfaction.
  2. jcsd
  3. Jan 29, 2004 #2
    ""Proof" therefore is essentially nothing save a series of concepts that explains reality to your satisfaction."

    There is joy and satisfaction that comes from choosing a "concept". And anything "new" often appears "wrong" at first. It is hard to have a new idea if everything we think is filtered by our past experience.

    I would like to chat with William James.

  4. Jan 31, 2004 #3
    James if great. But people have been saying that reality cannot be conceptualised for millenia. A concept is a metaphor, it can never be more.
  5. Jan 31, 2004 #4
    Proof means deduction from presumed assumptions and former results, or induction from statistical results on specific cases. So proofs are not final, only more or less convincing.

    But if James' pragmatic criteria were to be taken completely at face value, then psychologists and sociologists would be the only arbiters of scientific value. It pays to assume that specific truths can be approached, even if finality cannot be assured. It is also worthwhile continuing to persue alternatives even after open-and-shut results seem to be at hand.

    It is a wise saying: "Follow those who earnestly seek truth and retreat from those who claim they have it all."
  6. Feb 20, 2004 #5
    yes, you cannot prove anything. Once you "prove" something, it becomes a memory, and everything that exists could have been created an instant ago, and your "proof" is just a memory that could have been created. So it may not have existed at all. You can never know. Since you can never truly prove anything.
  7. Feb 28, 2004 #6

    Les Sleeth

    User Avatar
    Gold Member

    I think to understand James' statements about proofs one has to take into account the change philosphy was undergoing at the time. After centuries (millenia really) of a priori rationalization James, following in the footsteps of Locke and others, insisted reality is known not through reason, but through direct experience. In his Essay on Radical Empircism he writes, “Nothing shall be admitted as fact except what can be experienced at some definite time . . . everything real must be experienceable somewhere, and every kind of thing experienced must somewhere be real. "

    If we accept that experience is the most direct route to knowing reality, and reality is the "truth," then experience gives us truth while reason becomes the process of interpreting, calculating, and devising applications from what area of the truth (i.e., reality) has been experienced. Further, if one understands experience as one thing, and reason as another, then one must accept that each realm has its own methods of validation.

    The concept of "proof," then, is understood to be a validation process for the realm of reason. A proof is not required to give us truth but rather to provide a test for the logical interconnections and conclusions used in interpreting, as well as predicting where to look for, experience.
    Last edited: Feb 28, 2004
  8. Feb 29, 2004 #7
    Nicely put. With a bit of background added it would make a useful 'sticky'.
  9. Mar 1, 2004 #8
    Really? Prove it!

    Just kiddin'. My real question is, can you ever prove that you can't prove anything?

    In postulating that nothing is provable, you create a nasty paradox out of the simple fact that your statement cannot be proven by virtue of it's own correctness. Conversely, if you assume that the statement is not correct, then there must be some statements that are provable, and you are back where you started from.
  10. Mar 1, 2004 #9
    Anything can be proved or disproved given the right system of proof. But all such proofs are relative and therefore uncertain in absolute terms. This much is provable.
  11. Mar 2, 2004 #10
    There's no need for an absolute proof of anything...if you can get 99% of it right, you'll be ok, generally speaking.
  12. Mar 2, 2004 #11
    I suppose that depends how you define 'ok'.
  13. Mar 2, 2004 #12
    I define it as practical or useful. If 99% of the right thing will do the job as well as 100%, then why quibble over that 1%?
  14. Mar 2, 2004 #13


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Absolute vs. relative is not a difference in degree, but in type. It is possible in principle to be 100% right using a relative truth, provided you carefully consider the relations involved. Relative truth simply implies that a proposition's truth value is modulated with respect to some variable, whereas an absolute truth is constant with respect to all variables.
  15. Mar 2, 2004 #14

    Yes but if 99% of a proof will do the job then why not 98%?

    I agree that usually we have to make do. But surely usefulness is nothing to do with proof, or even with what is true.
    Last edited: Mar 2, 2004
  16. Mar 2, 2004 #15
    I agree. Where do you think that leaves something like the incompleteness theorems. They are true for all sets of variables. Does that make them absolute truths?
  17. Mar 2, 2004 #16
    Right, but I don't need to know what is "true", I need to know how to get from point A to point B, and hopefully sometime before the clock runs out. "Truth" that isn't practical is a useless notion, because it doesn't get you anywhere. Further, how do you know it is 100% true? While you go figure out how you go about confirming the truth about something, I'm gonna go live a little, ok?
  18. Mar 2, 2004 #17
    Fair enough. You are clearly content in the world of appearances. :smile:
  19. Mar 2, 2004 #18
    And you'll have to show that there is anything more to reality than that, you dig?
  20. Mar 2, 2004 #19


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You concede as much yourself when you say you get by on "99%" truth.
  21. Mar 2, 2004 #20
    Do I? It was just a random number I threw in there, you know...what I mean to say is that whenever I know enough to get from A to B, I feel like that is enough knowledge. Is there more knowledge out there? Quite possibly, but if it isn't going to do me any practical good, I don't sweat it much.

    For every question that we have an answer to, there are hundreds that we either haven't answered, can't answer as far as we know right now, or the answer doesn't satisfy someone.

    Gravity seems to be a favorite "whipping boy" for my sort of position. There are lots of theories of gravitation. Do we need to know which one is the "true" one, in order to make accurate predictions as to what effect gravity is going to have in 99% of situations we are going to encounter?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook