What is a scientific explanation?

Question

My definition of when something is "scientifically explained" is when the laws of science (a set of axioms) can be used to deduce behavior of said phenomenon quantitatively/qualitatively.

With this definition it seems that in principle, nothing is "unexplainable".

My reasoning is as follows: Suppose person A comes up with some phenomenon at some point of time that he claims is not explained by the science of the time. All scientists realize he's right. They have two options now to "explain it":

1. The standard approach: Tweak the laws a bit so that all that was previously "explained" still remains so (more or less) but this new phenomenon is also deducible from the new laws.
2. The desperate trick: Add this phenomenon with its quantitative details into their laws of science (axioms).

So, in principle this can continue forever. It seems like everything can be explained in principle.

The reason I ask this here at PSE is coz, to me, it opens up some philosophical questions:

1. Is there any essence to a "scientific explanation"? Does it mean anything at all for a phenomenon to be explained?
2. Can science ever "fail"? It seems like failure is ill-defined for science. Science can only either succeed or do nothing.
3. If all of this is true, it seems like there is a blur between "natural" and "supernatural" explanations to a phenomenon. We can consider a "scientific explanation" to also be a "supernatural" one, expect that the scientific explanation is more mathematical.

Answer 1

2.Can science ever "fail"? It seems like failure is ill-defined for science. Science can only either succeed or do nothing.

I suppose that science can fail locally (the Tacoma Narrows Bridge) or globally (the burning of the library at Alexandria). Also, I would say that science fails when scientists defy the common good and publish a bunch of dishonest crap for selfish reasons... preventing this is the whole point of peer review. But magazine interviews and TV documentaries don't go through peer review.

Still, I think you're right: with the right definition and limits, science as a methodology does not fail.

3.If all of this is true, it seems like there is a blur between "natural" and "supernatural" explanations to a phenomenon. We can consider a "scientific explanation" to also be a "supernatural" one, expect that the scientific explanation is more mathematical.

I would define nature as that which is repeatable and explainable without resorting to unexplainable things. An apple falling is natural, since no part of that is unexplained by natural causes: the biological apple and the physical principle of gravity, which in turn have explainable causes: biological plant life, physics of mass / astrophysics / gravity.

But I would assert that a person is not natural, partly on account of the originality (unpredictability) of our actions. The way we use the words "artificial", "un-natural", "fake", etc. bears out this assertion.

1.Is there any essence to a "scientific explanation"? Does it mean anything at all for a phenomenon to be explained?

I'm afraid this may be beyond my pay grade. What did you mean by "mean"? Your definition for explanation is pretty good, so apply that along with limits of science and the pitfalls of scientists, and you're probably there.

Answer 2

1) A philosophical explanation for an observed phenomenon with its structural basis in experimentally-supported scientific theories. So, the theory of the thunderstorm, from the water cycle and electrostatic discharge theories, which can be experimentally observed in a laboratory, is a scientific explanation of a thunderstorm. A different philosopher might say, "The Rain Deity sends clouds to us," but my belief as a philosopher of science is that neither of these theories are incompatible, nor do they invalidate each other.

2) Science "fails" frequently, because the human capability for comprehension of the natural world is bounded by the senses. Consider the number of unexplained aircraft disappearances and fuselage failures. To me, this is because although scientists understand the principles of flight, they do not yet understand the sky.

3) Because the human capability for comprehension of the natural world is incomplete, and bounded by the senses, even those natural phenomena crafted by the human mind (like flight, or the internet, or atomic technologies) assume a supernatural quality. I have transmitted this message to you an unknown distance at nearly the speed of light. The specifics are comprehensible, but is the fact that this remains possible in a world with such strife natural or supernatural? That is a matter of opinion. Lunar motion is, to some degree, chaotic and unknowable, but humankind landed on the moon in the middle of the Cold War. Again, you may ask, was that natural or supernatural? Again I have no answer.

Consider Von Braun: "Everything science has taught me strengthens my belief in the continuity of our spiritual existence after death." This was not an uncommon belief among the scientists of the early 20th century, regardless of who they served, and indeed the early alchemy practices which developed into a more-useful chemistry were almost indivisibly tied to the mystic practices of a variety of faiths.

Finally, I doubt that anything is "unexplainable," but, as we know from the quantum theory, certain things may be unknowable, and this would naturally lead to unsatisfying or unhelpful explanations.

Answer 3

It is only with a heavy heart that we resort to your Method 2 (assert the new empirically-observed truth to be an axiom). In some sense, we feel that an explanation which adds new axioms to the world is less successful than one which does not. This is the principle of Occam's Razor, by the way.

You can see this right now with general relativity and quantum mechanics. We know that the theories are both incomplete, because they are both incompatible. We have added them to our arsenal of engineering tools nonetheless, because they are more than complete enough to be very useful in their domains of influence. So here, "science" has been very successful, in that we have deduced some extremely handy facts about the way the universe works; and "science" has failed, in that we have no unified theory that works across all domains. We know enough for our knowledge to be useful, but not enough to be satisfied that our work is done.

Much of the advancement of knowledge is like this. Success is not a binary, but a continuum.