The philosophy of science

[Bluegoat]

I hope I've put this in the right place.

I was ready a rather predictable discussion on a forum the other day on creationism vs. evolution, (which is not what I want to discuss!) I noticed that there were many people that had a rather poor understanding of the scientific method, talking about theories being proven, what the method was supposed to do, and so on. This included all the actual science students, including grad students, in the discussion.

Now, I remember actually learning about the scientific method once in school, in grade 8. My teacher told us that a theory could never be proven, and I as a fairly bright 14 year old thought that was BS. I only understood as I matured, and really understood it more technically when I learned as a philosophy student about induction and deduction. But I always figured science students in university would spend a fair bit of time on this.

My husband, who studied chemistry, tells me he didn't touch the topic in university, it was assumed they learned it in junior high.

I'd be interested to know what others, especially teachers and scientists, think of this.

[postodave]

Well I am not trained as a philosopher or a scientist but I can tell you what my understanding of why a theory can never be proved is and you can correct me if I am wrong. for any inference that takes the form if A then B I cannot say because B therefore A. So I can never say if a theory was true it would have consequence x we observe x therefore the theory is true. However given if A then B we can say not B therefore not A. So it is possible in principle to falsify a theory but never to verify it unless we can directly observe the thing the theory is about. Something like this seems to have been the view of both Popper and Mill. However I have noted that Polanyi for whom I have great respect says a theory can be known to be true and I have never followed his argument well enough to see why.

[Kolbitar]

[Kolbitar]

BTW, greetings, BlueGoat...

[Bluegoat]

THanks for the greetings, Kolbitar.

What they teach these days is what I was wondering. Is it usual for scientists to not understand the nature of scientific knowledge? This is not, I'm sure, true of the good ones, but what of the masses that graduate from science programs in universities? If it is normal, what does that mean to their ability to interpret the results of experiments, or even to design experiments? Should we be teaching more about this in, say, high school or university programs?

I have often thought that public education should include a logic component in the 12 to 15 age range, would that help? Or am I seeing a problem that doesn't exist?

[postodave]

I'm convinced you are right that children need to learn these things at school. My daughter who is 16 was told by her physics teacher that all the theories they learn in physics have been proved. She raised the problem of induction using the illustration of Russell's chicken and he refused to discuss the matter and insisted on sticking to his point. Here in the UK we have a national curriculum which insists on putting knowledge in neat little boxes and we now have a generation of teachers who also grew up on the national curriculum. So my daughter is used to being told that she is asking questions that are not on the syllabus. However in English her teacher insists that to understand literature you need broad general knowledge and insists on going off topic to provide it. Naturally my daughter likes that teacher.

However you have to remember there are these popular writers with a particular drum to bang. So you get Richard Dawkins saying evolution is not just a theory it's a theory with evidence. To which I want to say, 'and your point is . . . ?' It is very clear to me, and I've explained my understanding of why above, that no amount of evidence can turn a theory into a fact unless we can more or less directly encounter the thing the theory is about. And yet is not the best theory we have the best because we think it is in some sense the truest? - that seems to be Polanyi's point.

A few years ago I read a piece by a politician arguing against the teaching of creationism in schools. He used John Stuart Mill to buttress his argument. Mill had said that the state should only intervene if people were being harmed and this politician argued that children were being harmed by being taught creationism alongside evolution and therefore this should be banned. I was sure he had misunderstood Mill and when I reread M ill I was certain of it. Mill argued not that the state should intervene whenever a person is harmed but whenever they are harmed in a way that violates their human rights. Therefore he argues for freedom in education. Parents can have their children taught what they wish as long as when tested the children come up to scratch. So it seemed to me Mill would have said by all means teach children creationism if that is what their parents wish but when they do their exams they had better understand the theory of evolution. And that was the key for me that they must understand the theory not believe it. In the same way Mill said there is no reason why an atheist should not teach a course on Christian evidences as long as he was not required to believe in those evidences. I concluded that a state dictated curriculum can easily lead to a retreat from liberty and that is exactly what has happened since Mill's day. I would also see a definite parallel between Mill and Popper in the way their political theories and philosophies of science dovetail together.

[Robert]

I was always confused about the role of Science in grade school. It wasn't until I began studying Philosophy had I learned how one needed to take a more serious look at scientific theory versus what is real. For just because something behaves in such a way, or more specifically how something is proven scientifically, is not the same as subjecting it to a truth test, or demonstrating that it is 'real.' For there is a distinction between being 'methodologically' true and 'ontologically' true.

The difference between methodology and ontology is quite simple. The former is attributing reality to elements of a theory in order to make it fit with reality. The latter is nothing more than scrutinizing the reality of theoretical elements' or elements of a theory. For instance, the concise definition of an atom as being a necessary component of scientific theory of any given physical object has changed over the years. This is due to the continuous improvement of previous theories. This does not mean that atoms are merely bits of a theory that do not exist. Rather, it means that our definition of an atom, or a building block of the physical world, is constantly in flux and is only 'methodologically' true.

However, an ontological term is scrutinized by a different set of standards; namely logic. For while they needn't be empirically verified-thus they are less rigidly scrutinized empirically-they are logically scrutinized-they must meet the demands of rigorous logical tests. So, the scientific method yields methodologically true definitions that are subject to change and yet are scrutinized by empirical methods, ontological definitions are subject to rigorous logical tests that demand precision in this area but not empirically.

[Kolbitar]

[Mornche Geddick]

In designing any experiment you have to include a control. That is, you have to have two almost identical samples running side by side in which one condition is different. If the inference (A then B) is true, then the A sample will exhibit B and the not-A sample will not. If neither exhibit B or both do or only not-A does, then the inference is wrong.

The reason it is difficult to prove a theory by experiment alone is that it may be difficult or impossible to arrange for a control. (You can't arrange for the sun to rotate around the earth.) That's why the evidence for evolution, for example, is mostly forensic rather than experimental except in the case of bacteria, which breed fast enough for evolution to be actually observed.

[Kolbitar]

[postodave]

I think it's helpful to distinguish between the problem of induction and the problem of hypothesis. The problem of induction is that the next observation could break your otherwise unbroken chain of confirmations. The problem of hypothesis is that however many observations your hypothesis explains there could always be another explanation for those observations that no one has yet thought of. Though I suppose these two are linked because as soon as you start getting observations that do not confirm your hypothesis you are forced to look for a new one that will cover both the original observations and the new anomalies.

[AllanS]

[AllanS]

double post

[postodave]

It sounds delightful when you put it like that Allan but I think I'll go with Einstein and suppose that the Lord is subtle but not malicious. Because whether it was the intention or not the effect would be deceptive and the artist would know that; deception can be a part of art and the creator could be that kind of artist but then he would not be the God of scripture. It's a huge jump from a guess about the wine at Canaan to a guess about the whole universe. So while it is possible, just as it is possible that the whole universe is ten minutes old, there is no reason for thinking it so. As for the simple that are supposed to believe this I take leave to doubt that they exist for this view is not simple at all but a piece of sophistry. Philip Gosse who first proposed this was an extremely complex man. The creationists sometimes seem to take this kind of view but actually tend to fall back on it after they have failed to prove some point using arguments from evidence. But it is a lovely conceit.

[AllanS]