ProphetofOccam
Paranormal Adept
Unfortunately for some skeptics, the scientific principles of probability in the Null Hypothesis are often overlooked. It is typical of them to water it down to suit their own bias in a given debate. This is where most skeptics fail because they know that if they accept the scientific principles of the Null Hypothesis rather than their own watered down and biased version, then evidence is no longer purely material. But the slope gets even slipperier. There is always a margin of error in every scientific experiment. No tolerance is perfect. No material 100% pure. No set of conditions is 100% duplicable. It varies widely from tolerances that are down to angstrom units to variables that are literally as unpredictable as the weather ( Meteorology ). Then there is medicine where anecdotal evidence is used as part of the evidence for determining a course of treatment. Then we move into psychology, the status of which is still under debate. In the end, when the situation is objectively examined, what we tend to think of a "solid science" is far from justified in holding itself up as the purveyors of certainty in truth except where pure logic ( like math ) is concerned ( but even that has more to do with philosophy ). What science can do is make reasonable claims for what the probable truths are based on empirical evidence ( evidence gained by the senses ).
I'm going to keep this reined in by using binary logic, just for the sake of the response. 0 is false, 1 is true.
I'd argue you're running a little broad with the concept of the null hypothesis, here. It generally applies to controlled experiments with a starting value of 0, speaking in binary abstracts (or, predicting for a negative). The causality for the state of the 0 can't be proven or disproven, but is observable and quantifiable, none the less. We can identify it as "o" -- we can identify it as null. Therefor, we can also identify its change from "0." we are also aware that it can change state. The only thing we can't assume is whether or not the lack of state change is genuine or, if genuine, otherwise meaningful. Its only function is to allow for the implied proof of a negative, not the scientific validation of a phenomenon in the absence of testable data -- the entire premise rests on the idea that the data is testable and that the hypothesis assumes for 0.
The absence of testable data is what I'm referring to. You can't test an eyewitness testimony, for instance. You can't take what you've heard and review it in anyway. You can either choose to believe it and use the information given to guide your search for something that can be reviewed (tested), or you can disregard it in light of other data that makes the testimony unlikely. In that case, you can state that you believe that the truth value of the testimony is 0, which is a null hypothesis, but even when you compare it to the data that lead to your null state hypothesis, you haven't resolved anything -- the state of the truthfulness can't change and thus can't be observed to change (leading to a solution). If you assume the truth value of the testimony is 1, which is a standard hypothesis (predicting a particular positive state change), all you can do is follow up. In this case, when do you decide that you have spent enough time attempting to replicate the reported conditions of a phenomenon, with no similar phenomenon taking place, before you decide that the 1 value has changed state to a 0? That determination is highly subjective, especially given that there would be so many unknown factors regarding an unknown phenomenon. In either case, your ability to even determine the value of a hypothesis (whether the truth value should be considered 1 or 0) is realistically impossible on any objective level. Much of the evidence, and the hypotheses (see our Alien physiology thread), for the paranormal is untestable in this way.
You can't debate something for which you can't even observe a state change, nor can you debate something for which you've not enough information to determine which state is which to begin with. What, exactly, would you even be debating at this point? The only thing left -- the subject summation of the nature of states, which is objectively arbitrary.