What is proof?

[bumab]

A lot has been made about the lack of proof in many debates around here and in other forums. But what is scientific proof? Does it require a mathematical formula?

Princeton web dictonary:
a formal series of statements showing that if one thing is true something else necessarily follows from it

Philosophy Text:
A formal demonstration of the validity of a deductive argument

Mine:
a measure of alcoholic strength expressed as an integer twice the percentage of alcohol present (by volume)

Wikipedia:
a rigorous, compelling argument, including:
a logical argument or a mathematical proof (see also proof theory).
a legal proof.
a large accumulation of scientific evidence

What do you guys think? Can a scientific proof (so often asked for) be derived from a preponderance of evidence, or do we require sound, logical mathmatics or something else to claim "proof" of so and so, or some theory or another?

[Fishteacher73]

Proof should be something that is objectively derrived and the results can be repteated by others.

[sanctus]

but after your definition Fish, there are only mathematical proofs, as math is the closest I can imagine we can come to objectiveness (and it isn't objective, at the very basis there are axioms which you can accept or not).

[Fishteacher73]

Possibly the answer should be a concept that works when cross-refernced. Such as a theory that "fits" in with the other theories or evidence from different POV. (Not competing theories, but theories in related fields).

[Tormod]

I second Fish here. Objective evidence is required (although there is always bias of some sort).

Also, like Fish says, scientific results must be repeatable - ie, if I do the same experiment as you, under the same circumstances, I must get the same results. If not, there is an anomaly (requiring the theory to be rewritten) or a falsification (requiring the theory to be reconsidered entirely).

[bumab]

Originally Posted by sanctus

...there are only mathematical proofs, as math is the closest I can imagine we can come to objectiveness (and it isn't objective, at the very basis there are axioms which you can accept or not).

That's what I was thinking. When faced with defining "proof," I can't think of anything that fits the common perception: Irrefutable truth. Sure, one can say "I have proof of such and such," but it's really only evidence.

Evidence can lie, though. If, for example, I flip a coin 100 times and it's always heads, is that proof that the coin only has a heads side? No, it's just chance. Repeatablility only builds a preponderance of evidence, not proof.

Often we say things like "show me," but that's debateable. Take any UFO abductee and they'd claim they saw litle green men. The BELIEVE they saw little green men. That's proof to most people- I saw it, it happened. I don't think our senses are especially good providers of proof, based on that logic. Like Descartes, who started from the premise that nothing he sensed could be trusted as real.

So I'm thinking proof is a mathamatical thing that is rarely achieved. So I was more looking for a practical definition... In science, we often call a preponderance of evidence (repeatability), as "proof," or rather a good basis for a theory. So where do you draw the line? I don't think the whole Creationism thing will go away because of this problem- they claim scientists don't have proof. And I don't think we do.

But we have a dearth of evidence, overwhelming in fact. But it's not proof. Does the proof concept get thrown around to much in science and scientific debates?

[Freethinker]

Originally Posted by bumab

That's what I was thinking. When faced with defining "proof," I can't think of anything that fits the common perception: Irrefutable truth. Sure, one can say "I have proof of such and such," but it's really only evidence.

Smacks of postmodernism.

[bumab]

Originally Posted by Freethinker

Smacks of postmodernism.

And?

You're right, it does... I still don't see an objection. I really am looking for functional definitions of proof from a scientist's perspective. If that perspective is a modernistic one, so be it. If's it's more suited for postmodernity, that's ok too.

[Aquagem]

Originally Posted by bumab

So I was more looking for a practical definition... In science, we often call a preponderance of evidence (repeatability), as "proof," or rather a good basis for a theory. So where do you draw the line? I don't think the whole Creationism thing will go away because of this problem- they claim scientists don't have proof. And I don't think we do.

I would just as soon see "proof" expunged from our everyday scientific lexicon. Creationists demand "proof" because they know we will say, "Of course I can't prove every single tiny little trivial point of evolution theory", at which they cry, "SEE? Gotcha! That's why you have to teach Itelligent Design!" "Proof" is a weapon, like a crude club wielded by a barbarian in a laboratory of test tubes and lasers.

While reading about "reason" not long ago, I found a reference to "abductive" reasoning, which in the 19th century enjoyed some popularity along with inductive and deductive reasoning. Abductive reasoning is, approximately, "using the best available evidence to arrive at the best (but always conditional) theory". The article went on to say that this fits scientific reasoning better than the other two in most cases, but that abductive had since been absorbed into inductive reasoning. I figure that happened so we wouldn't have to say, "Science is the process of abduction of truth".

NOTHING in science or any other discipline is PROVEN. (Even a mathematical proof is conditional. What do the three angles of a triangle add up to? The answer and the proof depend on whether you're talking plane or spherical trig. cf. non-Euclidian geometry, the irreducibly statistical propositions of chaos theory, etc.)

While this might sound like postmodernist palaver, it isn't, because there's another dimension to science that creationists try constantly to downplay. Scientific knowledge has solidity, robustness, because, together with scientific theory, it is built as an interlocking WEB of observation, fact, testing, and conceptualization. In that environment, proposed knowledge that doesn't fit with the whole stands out like a sore thumb and demands attention. If its hypothetical structure can be supported with new evidence, the whole web needs to shift, e.g., Einstein's famous prediction that Sun's gravitation would deflect starlight. When it did, we had some shifting to do, and lots of followup to solidify the position of relativity in the scientific world view. This is different from postmodernism because postmodernism implies that things are true only according to the model (concept, theory, paradigm) you're comparing them to, and those conceptual pictures are culturally determined and therefore equal in stature. Creationists love that. It just doesn't happen to be true.

I wish I could find Stephen Jay Gould's half-facetious definition of fact (I'll look around for it and post it here asap) -- something like, "a fact is something so firmly supported that it would irresponsible not to grant it our conditional approval". Much better approach than "proof", which I consider an anachronistic term.

And to respond to creationists, challenge them to construct any kind of an integrated web out of their "theory" that in any way fits the observable world, that is, anything that doesn't come down to, "Cause God did it, and that's REAL science!"

[Aquagem]

Here's the quote from Gould referenced in my last post:
http://www.quotationspage.com/quotes/Stephen_Jay_Gould/

In science, 'fact' can only mean 'confirmed to such a degree that it would be perverse to withhold provisional assent.' I suppose that apples might start to rise tomorrow, but the possibility does not merit equal time in physics classrooms.

StephenJayGould