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

Reasoning about reason

  1. Jan 7, 2009 #1
    How do we investigate reason, when the only tool of investigation we have is reason itself? If we want to test the laws of logic, then we must in the first place reason about them; and we must reason about the outcome of our tests. But then we would have had used the laws of logic in the process, and that will have defeated our goals (or so it seems?).

    How do we get out of this paradox? What is the starting point?
  2. jcsd
  3. Jan 7, 2009 #2
    We call it reason because its reasonable.
  4. Jan 7, 2009 #3

    Math Is Hard

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think there are 2 regressing chains you could play with here:

    1) How can we ultimately know anything with certainty about our logic which leads to "how do we know with certainty anything about anything"?

    2) What is the general starting point of our thoughts about our thoughts about our thoughts (etc.) that we have, or where is the "top layer" of metacognition?

    For 1) I wouldn't have a clue. It seems to be sort of a dead end street that ends at cogito ergo sum.

    For 2) I imagine it begins in childhood. We experience regularities in the world and begin to test out theories and find that we can use logic to make consistent predictions. From that we can build a set of beliefs about how things work, and this becomes a framework that we can use to examine our reasoning about events metacognitively.

    We're sometimes wrong of course. Cognitive psychologists are very interested in examining how we reason and the errors and biases to which we are prone.
    Last edited: Jan 7, 2009
  5. Jan 8, 2009 #4


    User Avatar

    Interesting question. This is related to what's called "the problem of infinitie regression" that seems to appear in various forms of induction. Thus it's closely related to the philosophical "problem of induction". It's usually noted that an attempt to justify induction by induction leads to a problem of infinite regression.

    As I see it, there is a close relation between infinite regression and evolution. Most philosophers, like Popper has viewed this as a problem. I don't see it that way. It might be that infinite regress, is in fact a reasonable abstraction because it does after all seem like evolution in nature is almost like an infinite regression.

    This is related to things I'm currently researching myself. The feature that should distinguish infinite regress from fruitless circular reasoning is that there is an infinitesimal progress in the infinite regression, I think this can be related to evolution and self-organisation.

    I think most of those in that past who talked about infinite regress in that past in the sense that "see look what happens", isn't focusing on evolving processes, there were focused on capturing the snapshort of truths. Maybe the difficulty in doing that, is related to a flawed focus. This is what I think.

    My personal take on this, is to see that it really isn't a paradox, not more than the failure of finding a static solution to Einsteins field equations appeard to him was a paradox :) I used that analogy before because I think it's illustrates a point.

    About the starting point, in a certain sense I like to think that ANY starting point must be possible. But on a more practical level, that starting points I have personally chosen, is to combined the freedom of choosing initial condition, with the constrains of representing one, but a real observer or system. As the complexity scales down, the freedom is reduced by constraining the combinatorics. This is the starting point i've used ot make it easier, and then see how complexity instead can grow spontaneously. I have persoanlly drawn parallells to this also to the generation and origin of mass. It has similarities to the "infinite regress" thing IMHO at least.

    I've personally started to search for an abstraction, and I've podered the notion of a microstructure, where each microstructure has a complexity number or mass. Then the only choices are the microstate. By infinite regress against an unknown, the microstructure can increase it's mass by "learning about it's environment". The increase in effective information CAPACITY, vs information, is closely related to "generation of mass without mass".

    The original question is one philosophical dimension of a problem that has interesting counterparts also in physics.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook