Gah, so much to respond to in this thread.
I'll get to it, eventually...
No, relative truth is not a coherent concept. Truth is boolean by nature. Still, you're trying to equate our knowledge of the world with only specific true statements. Like, the following can be a true statement: 9999999999/10000000000 B follows A.
Furthermore saying something is true 99% of the time is not the same as saying I am 99% certain that it is true (the first is not possible as you pointed out, but the second is perfectly plausible).
Along your line of reasoning the entire idea of statistics can not exist...
How is this a wall for science? If we believe in the scientific method we can't observe the world around us?! *confused* <--- I assume this is not what you mean. Care to explain it?
No, it's not a valid conclusion. But we're perfectly capable of saying we know absolutely that 9999999999/10000000000 B follows A. We're more interested in working with what we can say rather then what we can't say.
You're sort of contradicting yourself (sort of, bear with me here). You say "
This is not attack against science based on its inability to attain 100% certainty." Then you go on to attack induction and causality, but induction is a line of reasoning which "supports" its conclusions, but does not ensure its truth (again, probability comes to mind, a conclusion with a whole lot of support is more likely (or should I say our certainty of it is more likely) true then a conclusion which has no support or support against it even).
And causality is a self-evident truth. We're not debating whether causality is true or false, whether some A definitely caused some B. We're debating if we can prove a particular A caused a particular B that we have identified.
-blazed
I don't think you understand his issue with induction.
Induction is inherently non-rigorous, by it's very nature it cannot account for the infinite potential non-observed phenomenon.
Your 9999999999/10000000000 times B follows A cannot, by it's very nature account for the 10000000000000000000000 times B did not follow A that you have not observed.
Yes, this does dispute the validity of statistics, but there's a reason why induction is not used in mathematics (the reasoning type, mathematical induction is actually a form of deduction), it cannot prove "for all", it just proves "for the sample".
The philosophy behind science however, is built to account for this issue.
The simple fact is that science does not hold that anything is absolutely proven, the theoretical underpinning of science is the idea of disproving the null hypothesis.
The null is simply "nothing" (whatever "nothing" is relevant, ex all swans are no color), which is in turn replaced by another null hypothesis when a contradiction is observed. That is in turn replaced by another null when a new contradiction is observed.
Take for instance, the statement "all swans are white". It was created inductively when somebody observed swans and only found white ones. Thus the conclusion was that all swans were white which became the null hypothesis.
However, this null was disproved when somebody discovered a black swan. Thus the null was rejected a new null hypothesis was created, "all swans are white or black". If a yellow swan is discovered then the same thing will happen once again.
This is the only was induction can be rigorous, infinite observations. If one observes all possible cases then induction becomes rigorous and therefore valid. Of course, that will never occur, so the net effect is that science is gradually moving closer to the truth by gradually rejecting null as new data contradicts it.
Understand? To deal with the problem of induction being non-valid, science rejected the idea of having the absolute truth in a finite time frame. That's why there are no more "laws of science", ANYTHING can be rejected. Like the law of conservation of energy and the law of conservation of matter were rejected.
PS. Statistics is included in this.
No clue. I personally doubt that there is a better system.
Great commentary on the flaws of science!
The problem is that you are a couple of decades out of date.
The scientific method uses empirical falsification, which is ultimately induction, is it's only rigorous form. It recognizes that induction for any finite number of tests is inherently invalid and therefore continuously looks to reject previous inductive conclusions with new evidence, Ultimately this provides infinite tests which result in induction being rigorous.
Of course, since infinite tests are ultimately impossible over a finite timeframe, this means that science recognizes that any and all of it's conclusions are ALWAYS open to being rejected due to new evidence.