View Full Version : What's Proof?

Eieam Wun
2005-Mar-19, 08:07 PM
I ask because I was recently discussing topics about proof and what it was which means convincingly demonstrated. But in the discussion was told that science "has moved away" from proof. If so what then does science strive to achieve for if not to prove their theories? Maybe it is evidence to show their theories are correct, or at least convincing? Perhaps it's to demonstrate their theories are sound? Just wondering aloud here and hopes for a bite


2005-Mar-20, 11:44 PM

You might want to visit a good philosophy site for this one, especially philosophy of science.

I know of two ways to look at "proof":

(a) Absolutely true in the sense that "there is no way this can possibly be wrong"
(B) True in the sense of "we are as sure as we can possibly be that this is indeed true".

Science is such that proof in the sense of (a) can never be achieved. No matter how "obviously true" some statement might be, there's always the possibility that something will come along and rip our precious theories to shreds, or at least supplement them (Einstein's various theories did the same to Newtons, especially in gravitational theory. It's not that Newton was just wrong through and through: it's that his theory was incomplete, though perfectly adequate for the time (the time's science, detection technology, predictions of planets/body positions of other bodies due to gravitational interactions, and so forth).

Newton was certainly true in the second sense up to the turn of the last century or so. The problem was that the technology and techniques of the time did not really allow for us to adequately challenge Newton. However, throughout much of the 20th century and up to today, most of Einstein's theories and conjectures have proven a more accurate description of how the universe works when higher levels of precision are required.

Of course there's always the question of the quantity and quality of evidence to support your view, the philosophical system you use to look at the evidence, and so forth. I don't know if English Common Law is the basis for your country's laws, but in the countries where it is the basis for law, there are three levels of evidence/standards of proof that courts can require, depending upon the kind of cases that are being tried

(a) Perponderance of the Evidence - this is the lowest standard of proof, usually used in civil suits. However subjective this may be, this level of evidence merely states that the plantiff must merely convince the judge that the evidence is "more likely" to be on the plantiff's side than against it. (for example, a suit concerning bodily injury on someone's private property - assuming the injury is NOT of literally criminal negligence)

(B) Clear and convincing evidence - this is used in trials surrounding claims so out of the ordinary that considerable evidence need be present (for example, a legally enforceable contract signed under duress).

© Beyond reasonable doubt - this is the standard used in criminal trials, since the freedom (and in some cases, the life) of the person is at stake. In this case, not only does the state's attorney have to make a good argument that the accused did do "it", he or she would have to defeat ALL arguments put forth by the defending lawyer that the accused did not do "it".

Truthfully, this all but exhausts my knowledge of the matter. There are literally hundreds of books out there surrounding the issue of "proof". I hope this helps

2005-Mar-21, 12:30 AM
:huh: . From an elderly school teacher. ( so it must be right )
If you arrive at a conclousion which gives rise to a theory, your next task is to disprove your theory. Only then does your findings become a lagitamate explanation. How much effort you apply to this task is the herdle. For some a life time. While for others? they never lower themselves to admit error.
I'm in the first group. I am often mistaken. :rolleyes:

2005-Mar-21, 01:11 AM
You might want to visit a good philosophy site for this one, especially philosophy of science.
The big names include Popper, Kuhn, Lakatos, and Feyerabend. This site (http://www.eblaforum.org/main/viewforum.php?f=11&sid=1744b9e037e0d396a5bd5dd2505f6fa0) is a discussion forum on HPS, and the materials Hugo lists in Suggested Reading and Resources include the landmark works of the big names, as well as some introductory books (I agree with Hugo, the Chalmers book is good; you might also enjoy Robin Dunbar's "The Trouble with Science").

Eieam Wun
2005-Mar-21, 02:40 AM
Very thought provking reply filrabat, well appreciated.