What can science prove, and not prove?

[unkleE]

I want this discussion (if anyone replies) to be on the issues I raise please, and not just become a Dawkins bash - I am mentioning him simply to introduce the issue.

Richard Dawkins has said many times that there is no scientific evidence for God's existence, therefore we can safely assume God doesn't exist. One example is mentioned in this article, where Dawkins is reported as saying that "the existence of God was a scientific issue. If, in effect, God could not live up to the standards of scientific proof, then He must be declared dead." I have seen him say similar on video, and seen him reported as saying similar as well, so it isn't an isolated or unfair quote.

Like I said, I don't want to bash, or even mention Dawkins from here on, I just want to discuss the issue: what can science prove and what can it not? I want to ask a few questions and see what people here think please:

1. When we talk of a scientific issue, I presume we are talking about the process of developing a hypothesis and the means to test it, then running the tests, doing the stats on the results and coming to a conclusion, which may then re-shape our hypothesis?

2. I presume we agree that one cannot "prove" (within reasonable confidence limits) God's existence scientifically in this manner?

3. So now the interesting question: what important things can we or can't we "prove" this way? Here's a few thoughts:

• Almost everyone would agree that pedophilia or the Holocaust were/are ethically evil, but is there any way we could demonstrate that scientifically in the way outlined above? I understand Sam Harris, and utilitarians, think we can, but what I think they show is that we can compute benefit and harm and use them to assess actions, but not show that they are wrong. Do you agree?
  
  
• Questions of ancient history don't seem amenable to this hard-core scientific approach either. Even the existence of Julius Caesar, which I presume is not in doubt historically, could not be assessed by the scientific method, except in the most general sense that would allow us to assess at least the existence of Jesus. Is this enough to meet the requirements?
  
  
• From my reading of evolutionary science (admittedly very cursory) I understand that we cannot prove abiogenesis and some aspects of early biological evolution in the scientific sense outlined above, but rely on feasibility studies and other indirect methods.
  
  
• I wonder whether the multiverse can be, or can ever hope to be, demonstrated by hard science. My reading of people like Susskind and Rees suggests we cannot, and probably never will be able to.
  
  
• Obviously we form relationships, get married, trust our parents, etc, without anything like scientific "proof", but is this example a valid one for this discussion?
  
  
• Political science is another example - politicians take courses of action to address known problems, and we choose which ones to support, but our level of certainty is way lower than scientific demonstration, but, again, is this a useful example???
  

So there's a few ideas. I'm sure these matters are much discussed among philosophers, especially philosophers of science. Has anyone any wisdom to offer please?

Thanks.

[Mike D]

The oft-repeated mantra on what is and what is not scientific seems to involve 3 words (or variants thereof):

Observable
Repeatable
Falsifiable

Now I can see the reasoning behind these points, but I think the use of them in conjunction shows a kind of confusion, mostly because the first 2 deal with the observational/experimental side of science and the third deals with the theoretical side: you can only falsify a theory, you can't falsify what someone has observed; and conversely, you can only observe or repeat a phenomenon or experiment.

Taking the first 2 - observable and repeatable - it seems like a truism to say that science can only deal with the phenomenological and then only insofar as it is a regular, repeatable phenomenon. So for example, anything internal to my mind and experience (at least currently - maybe neuroscience will read minds and emotions at some point, though I ain't holding my breath) and anything unique are immediately ruled out of bounds. Looking at the list in the OP, I think many of these would be immediately ruled to be outside the scope of science because either they are not objectively observable and/or repeatable.

My thoughts on this:

1. Reason and science are not the same. Okay, it's a statement of the bleedin' obvious, but in light of a lot of material (from both believers and atheists) I think it needs stating. You might think the basis of your morality (or whatever) is reasonable, but that doesn't make it scientific (nor should it need to be). However, if it is taken as an article of faith that only scientific knowledge (so called) is valid, then you get a lot of things that folks attempt to shoehorn into a scientific framework and that really don't fit.

2. Speaking as a statistician: can science actually 'prove' anything, even those things which can be objectively measured? My answer would be no, and in fact the whole falsifiable criterion is based on this - you can't prove anything with science, you can only disprove. At the best, evidence can be 'consistent with' a hypothesis. So does this make the aforementioned position that only scientific knowledge is valid an epistemologically nihilistic position i.e. you can't actually 'know' anything? *somewhat tongue in cheek smiley*