I think science, like many other things, is a family of separate enterprises that resemble each other without strict transitivity of similarities. There is no essence of science; there are properties that some scientific enterprises have that other scientific enterprises do not.

What happens, I think, is that an ideal image of science is covertly substituted for the actual manifestation of science. The "scientific method" is an example of this. There is no one single methodology shared by all the sciences, nor is the "proper", "tried and true" methodology always followed. There is no single scientific "culture". There is no shared belief system across all scientists. There is no shared qualities of scientific theories that make them specifically scientific.

Lots of scientists have opinions about what they think science should be. But what a scientist says about a scientific topic is far different than what they say about science as a field, which is a philosophical, normative topic. I think this distinction gets overlooked sometimes - just because a person is a scientist does not mean they have a good understanding of what it is they are doing or what it represents.

I'm confused here. I thought it was well understood that science is rooted in metaphysical naturalism and epistemological empiricism. Empirical testing is what underlies any scientific endeavor - you have a question about the world, make a prediction about how something in the world behaves, and test to see if it reproducibly behaves that way. This is the method common to psychologists, physicists, biologists.

I don't really get the impression that there is an overriding method or metaphysics in science other than what people pay lip service to. — Andrew4Handel

Your impression is mistaken, interest in the topic has been lively for several centuries and there's a whole branch of philosophy dedicated to it, called "philosophy of science," with several major theories which tend to be roughly similar in overall outline, though they differ in details, both in terms of epistemological and metaphysical subtleties, and in terms of special conditions and limitations necessary for particular sciences. For example, there are strict limits on what's possible in the way of experimenting with human beings, which limits the kinds of investigations that can be done in psychology and the social sciences as compared with, say, physics or chemistry; or again, although history is in some ways a scientific discipline, the fact that you can't rerun historical events and change the parameters severely limits how scientific history can be.

The general idea with science is that it's both a co-operative and competitive enterprise involving the generation of informed guesses (hypotheses) whose concrete, experiencable implications are deduced and then tested, both in themselves and as compared with what would likely eventuate if the hypothesis is false. Scientists publish the results of their tests and the methods they used, so that other scientists can try to either confirm or disconfirm the results of their testing.

That's the basic logical part of science, but there are also both overall traditions that scientists follow with regard to science in general, and specific learned traditions (things that you learn from experience, tricks of the trade so to speak) for the special sciences in particular, that go along with that basic schema. For example, in order to generate a pool of good informed guesses scientists often do a fair bit of data gathering and observation. But the crucial bit is the deduction and testing - the deductions have to be precise and logical, and the testing has to be as rigorous as possible in order to isolate the causal factors under consideration as much as possible.

I don't think science has to make any metaphysical commitment like naturalism. I think naturalism and physicalism are quite meaningless in terms of picking out entities. But is science wants to promote a metaphysics it would then become a philosophical competitor.

I don't think you have to ask questions to make scientific discoveries. Just mixing chemicals can create innovations and new phenomena can appear under a microscope or any other form of observation..

I think most claims in evolution and cosmology are unfalsifiable because they are historical claims about one off events that can't be replicated.

What people must have used in the millennia before the scientific method was developed was practical problem solving.

For instance, [it is thought that] hunter-gatherers harvested kernels of grain and learned that the kernels had to be scattered around to get more of that food later on. They may have discovered that using bigger kernels of grain led to more big kernels of grain being produced. At a later date they apparently discovered that grain that fell on hard soil didn't grow. From this insight hunter-gatherer-agriculturalists starting roughing up the dirt a bit with a stick. Eventually (over what... maybe several hundred seasons?) a system was worked out to grow reliable crops of grain.

There was no theory and practice didn't become theory.

There are many early innovations that had to proceed from luck, observation, and some simple problem solving. Someone, for instance, discovered that nuggets of metal could be worked and turned into metal versions of bone or stone tools. Later on it was discovered that metal could be melted; still later it was found that heating rocks would melt out metal. Smelting metal took many centuries of practical effort to emerge.

Using practical methods led to many critical innovations over several millennia, quite a few of which are still in use.

I don't know of any evidence to suggest that pre-modern people formulated a "system of practical common sense" which could fuel a revolution. The consequence is that technological development was fairly slow and tended to hit fairly low ceilings.

The scientific method is, among other things, a way of turning practice into theory and turning theory into practice.

What I am thinking is that science might be just a very diverse range of practises with no underlying metaphysical claim to be found or to unite it.

But people present it as worldview or the only method of truth or as a reductive process.

I think it is too restrictive to try and reduce it all to physics or the physical or empiricism and neglect the role of the imagination, cognition,chance, invention, intuition, desire, bias, political forces, commercialism and so on.

What I am thinking is that science might be just a very diverse range of practises with no underlying metaphysical claim to be found or to unite it. — Andrew4Handel

Why not begin by listing all the things science doesn't do then - like reading goat entrails, or accepting personal proclamations of faith, or wasting too much time on untestable speculation?

Do you think that a story about metaphysical naturalism and epistemological empiricism won't emerge, exactly as @aporiap says?

And maybe the concrete uniting feature is that science is generally pragmatic - not so fussed about prescriptive methods as you seem to think it ought to be. There are recognised good habits - well explored in philosophy of science - but also plenty of room for science as a creative art. It is not paradoxical that a method so powerful can also afford to be quite relaxed in many respects. It works so well that it can be sloppy in some regards.

That is why philosophy of science talks of paradigms and the difficulty shifting them. Science often tends to assimilate evidence to existing wisdom rather than being built up of discoveries, paper by paper.

So it is clear to anyone that science is generally committed to metaphysical naturalism, and then has a particular affection for atomism within that. It is also generally united by a method of reasoning that involves the cycle of abductive hypothesis, deductive theorising, and inductive confirmation. To deny this, as you do, is just unreasonable.

I think it is too restrictive to try and reduce it all to physics or the physical or empiricism and neglect the role of the imagination, cognition,chance, invention, intuition, desire, bias, political forces, commercialism and so on. — Andrew4Handel

But science is also pragmatic and so doesn't believe that it needs to stick to some rigid approach. It is pretty flexible within the general limits that define it. There is room enough inside for quite a variety of ways of attempting to achieve progress.

^^

I don't think science has to make any metaphysical commitment like naturalism. I think naturalism and physicalism are quite meaningless in terms of picking out entities. But is science wants to promote a metaphysics it would then become a philosophical competitor.

Well epistemologically it's wedded to empiricism which makes it very partial toward restricting what 'is' to what can be seen and felt. There are anecdotal accounts of supernatural experiences but they don't meet the reproducibility criteria for empirical facts, the accounts aren't consistent with each other and simply can't be tested or explored.

I don't think you have to ask questions to make scientific discoveries. Just mixing chemicals can create innovations and new phenomena can appear under a microscope or any other form of observation..

I think most claims in evolution and cosmology are unfalsifiable because they are historical claims about one off events that can't be replicated.

Hmm so the critical, defining feature of empirical testing is more so the confirmation of reproducibility for a given theory. It's true penicillin was accidentally found to kill bacteria in an agar plate, but what solidified its credibility as a theorized antibiotic was the subsequent testing of and confirmation of its killing capacity. Accidents can play a role in scientific progress but empiricism is always necessary to confirm the hunch.

Also, evolutionary and cosmological theories are falsifiable, they make testable predictions. While they they make inferences about how things were in the past, those claims logically follow from the underpinning theories which themselves are testable.

I think metaphysics and science have the same origin - Philosophy. Both metaphysics and science went hand in hand, each in its own place and with its own utility. This was the case in the classical period of the likes of Pythagoras, Hippocrates, Democritus, Plato, Aristotle, etc., and later with Leonardo da Vinci, Newton, Swedenborg, Einstein, etc. Newton's Philosophiæ Naturalis Principia Mathematica (Mathematical Principles of Natural Philosophy) is a good example of it.
Today, philosophy is being restricted to pure logic while metaphysics is being discarded as nonsense. And while science's basis of empiricism is admirable, we know there is more to experience than the physical or tangible. Hence, metaphysical themes are returning under the guise of science, e.g., psychology, acupuncture, etc. This is because, even though we cannot categorize them as tangible or physical we understand that, empiricism, being limited to a person's capacity, is not comprehensive enough and, therefore, as long as we can reflect on effects and deduce a system/methodology of application, then they may as well be included in science, e.g., the modern theory of dark matter and dark energy which is a modern reprisal of the ancient theory of aether.

I don't think science has to make any metaphysical commitment like naturalism. I think naturalism and physicalism are quite meaningless in terms of picking out entities. — Andrew4Handel

Comes to mind the scientist who says he/she has absolutely no philosophical views or any interest in philosophy: He/she just runs the tests, uses simple statistics and that's it.

Also, evolutionary and cosmological theories are falsifiable, they make testable predictions. — aporiap

There are a huge range of claims made in evolution beyond the general paradigm. I am thinking about individual claims as opposed to support for the general paradigm..

For example claims about homology in fossils. Homology claims about humans and other living organism can be disproven by DNA testing so that we find that Tom Cruise look a likes are not usually related to him and similar looking animals can be unrelated.

The point I am making though is that science doesn't always demand empiricism and falsifiability (see string theory, many worlds the multiverse)

Empirical evidence might prove or falsify a theory but science is not constrained or restrained from speculating.

I think some of the prominent metaphysics and methods used in science have made the study of mental states nearly impossible because of their invisibility and subjectivity. For example people face a lot of skepticism when they have a hidden illness or mental illness, Some claims just have to be taken on faith and some inferred. Among these things are chronic pain with no visible bodily damage or abnormalities, phantom limbs, hallucination, M.E. etc.

I am not saying claims have to be falsified to be true but just that this is not a uniting principle in science. If it was then the burden of proof would probably become too high.

Statistics are an interesting example.It is quite easy to collect data but interpreting it is a different matter.

There are flawed, biased political and simply incorrect interpretations of data.

I think the contrast between science and the supernatural is misleading.

For example think of the unimaginable evil humans have perpetrated like the holocaust.

I don't think anything credited to spirits and demons has matched the actually reality of barbaric human conduct but can science seriously reduce this to neuronal firings?

I think science lacks the qualia of reality that is the vivid experiences including immense suffering, prejudice, infatuation, music and so on.

The danger of materialism is extreme objectivism just treating people like insentient objects to be mathematically manipulated.