Scientific testing (split from goal of science)

[Reg Prescott]

@ - "Evolution is consistent with what we observe"

 

So you think it might be true? Or have a chance of being true? Or somehow truth-like? (Or even entirely untrue?)

 

If not, why are scientists invariably appalled when our religious loonie friends refuse to accept it?

Edited October 2, 2015 by Reg Prescott

[Strange]

If not, why are scientists invariably appalled when our religious loonie friends refuse to accept it?

 

Because they are denying the evidence simply because of their personal beliefs. That is the antithesis of science.

 

But perhaps you have put your finger on part of the reason for the problem creationists have with science. They think their beliefs are "true" in some absolute sense (independent of any evidence). Whereas, the scientific definition of "truth" is more like "consistent with what we observe" (and, for realists, this will be extended to "consistent with reality"; I won't mention idealists as it might annoy you).

[Reg Prescott]

Strange - "Whereas, the scientific definition of "truth" is more like "consistent with what we observe" (and, for realists, this will be extended to "consistent with reality";

 

 

But you've been telling us over and over again science has (almost) nothing to do with truth.

 

A change of heart?

[Klaynos]

@ - "Evolution is consistent with what we observe"

 

So you think it might be true? Or have a chance of being true? Or somehow truth-like? (Or even entirely untrue?)

 

If not, why are scientists invariably appalled when our religious loonie friends refuse to accept it?

Can you define true? It is consistent with the observations. It is the most constituent theory with those observations that we have.

[Strange]

Strange - "Whereas, the scientific definition of "truth" is more like "consistent with what we observe" (and, for realists, this will be extended to "consistent with reality";

 

 

But you've been telling us over and over again science has (almost) nothing to do with truth.

 

That's right.

 

The reason that I put the word "truth" in scare quotes is because the word is rarely used in science. If it is used in popular science writing or journalism, then it is used in (roughly) the sense I suggested. This is very different from the commonly accepted sense of truth and certainly from the religious meaning. Most people would take it to mean something that is unchanging and irrefutable. If it is true today then it must be true tomorrow, and so on. (You hadn't defined what you mean by the word, so I had to guess.)

 

Science is not like that.

 

Scientific theories are not regarded as being "true". At best, they are provisionally correct for a limited domain of applicability, according to our current knowledge, within certain error bounds, etc.

 

All the above caveats are (or should be) implicit in any discussion of science. They should not need tacking on to the end of every single statement.

 

Science is also able to have multiple valid, but inconsistent, theories at the same time. Newtonian gravity and GR cannot both be "true" and yet they are both accpeted, mainstream, useful theories that are used every day.

 

I really don't know what the point of this thread is any more. I think everyone agrees that science is not done in the caricature process you described at the beginning. So what else is there to say? You seem to have moved on to whether theories describe reality or are just models. But that was the topic of the thread this was split from. Perhaps you should take the discussion back there?

Edited October 2, 2015 by Strange

[ajb]

But you've been telling us over and over again science has (almost) nothing to do with truth.

Maybe you should define 'truth' and therefore avoid this round-a-bout arguments.

 

As Strange says, the closest to 'truth' in the scientific context is that our models/ideas etc match what we observe to a reasonable degree of accuracy. Any deeper meaning has to in the domain of philosophy.

[Reg Prescott]

@ Klaynos

 

I'm simply using the term "true" in what I take to be its everyday usage (what philosophers refer to as "the correspondence theory of truth")

 

A statement is true if it corresponds to actual states of affairs in the world. E.g.

 

The statement "The dog is drinking water" is true if and only if the dog is drinking water.

 

The statement "All copper conducts electricity" is true if and only if it is indeed the case that all copper conducts electricity.

 

The statement "Humans evolved from apes (or apelike creatures or whatever)" is true if and only if this statement corresponds with what is the case in reality.

 

It's not the only theory of truth out there, but I assume it's the one we all use on a daily basis. Or do you have another in mind?

So do you believe science should AIM TO produce statements/hypotheses/theories of that type?

 

Or AIM TO produce statements that are false?

Edited October 2, 2015 by Reg Prescott

[ajb]

So do you believe science should AIM TO produce statements/hypotheses/theories of that type?

 

Or AIM TO produce statements that are false?

I am going to say no.

 

The statement that 'if I drop a lead ball it falls to the ground' is true. However, this statement is not enough to be a physical theory. I need a mathematical framework to make predictions such as how long it will take the ball to hit the ground from a given height, how fast the ball will be travelling just before it hits the ground and so on. I can then compare these numerical predictions with observations.

 

In other sciences it may not be so clear cut as this, while in mathematics the aim is to try to produce only true statements!

[Reg Prescott]

ajb -- "In other sciences it may not be so clear cut as this, while in mathematics the aim is to try to produce only true statements"

 

Well, why wouldn't you want that for yourself in science? Even if you can't always attain it, don't you think you should AIM for it?

 

Or do you think scientists should deliberately aim for a few false statements here and there?

 

* raises another eyebrow *

[Strange]

So do you believe science should AIM TO produce statements/hypotheses/theories of that type?

 

Or AIM TO produce statements that are false?

 

How about "neither"? Does "neither" work for you?

 

But, seriously, ajb is right: it is not just about true or false statements about reality (those may be evidence used to support or refute a theory) it is about explaining those facts. Or, better, creating explanatory models.

 

So, evolution happens. That is a true fact (like "gravity exists"). But the theory of evolution is a model (or a set of models) that attempt to explain the processes by which this happens. And asking whether this theory is "true" is, as the quote you posted earlier pointed out, fairly meaningless because it depends on exactly which part of the model you are talking about. At the broadest level of "population variation is acted on by natural selection" I doubt anyone (rational) would question it. However, once you get into the details of exactly which traits are inherited and how, or which factors are selective, what the relative importance of different sources of variation are, and what other factors may be involved then it becomes increasingly less clear cut as to which bits are consistent with the evidence ("true", if you insist) and which are not.

 

 

Or do you think scientists should deliberately aim for a few false statements here and there?

 

That is just another silly strawman argument.

Edited October 2, 2015 by Strange

[Klaynos]

@ Klaynos

 

I'm simply using the term "true" in what I take to be its everyday usage (what philosophers refer to as "the correspondence theory of truth")

 

A statement is true if it corresponds to actual states of affairs in the world. E.g.

 

The statement "The dog is drinking water" is true if and only if the dog is drinking water.

 

How have you tested the water for impurities? Can you prove that what the dog is drinking is 100% H2O? If you cannot then how is that statement "true"?

 

 

This is also an observation, not a prediction.

Without knowing how to test trueness, without a clear definition of true your argument just seems and argument for the sake of it.

[ajb]

Well, why wouldn't you want that for yourself in science? Even if you can't always attain it, don't you think you should AIM for it?

 

 

Or do you think scientists should deliberately aim for a few false statements here and there?

You can always turn any question into one of asking is a statement is true or false. That does not mean you can always clearly decide.

 

For example, is the statement 'the acceleration due to gravity at the Earth's surface is 9.80665 m/s^2' true or false? How would you decide?

 

Well, I claim the best you can do is preform some experiments, take into account experimental errors and so on and see that the above statement is consistent with what one observes.

[Reg Prescott]

Klaynos -- "How have you tested the water for impurities? Can you prove that what the dog is drinking is 100% H2O? If you cannot then how is that statement "true"? "

 

A key question indeed, and one that has been causing all kinds of confusion throughout the thread. Take a statement such as:

 

"The number of stars is odd"

 

Presumably this statement is either true or false (ignoring unnecessary complications). Would you agree?

 

Now, whether we can KNOW (or prove it) is an entirely different question, an epistemic question. Do you see the difference?

 

Now do you see the problem with your statement above? Presumably the statement "The dog is drinking pure H20" is either true or false. Agreed?

 

Whether we can KNOW this is an entirely different matter.

or if this is more clear.... the fact that we do not, or cannot, know which value (T or F) the statement should be assigned does not imply that it has no determinate value.

 

Do you see?

Edited October 2, 2015 by Reg Prescott

[ajb]

"The number of stars is odd"

In science one would be looking to make a prediction that the number of stars is odd and not even, say in our galaxy. In principle, mod complications of the bounds of our galaxy and what counts and does not count as a star, one could count the stars and decide.

 

If it just happens by chance that the number of stars is odd and that this is not tied to some theory, then the results is superficial. Science is not just about making true or false statements, even if we can clearly decide if a statement is true or not. (For this example I am not convinced it is a clear question given stellar evolution and deciding on boundaries )

[Reg Prescott]

I did say "avoid unnecessary complications. Ok, try this then...

 

Say, there is a certain number of marbles in an urn. We do not, or cannot, know what the answer is (we're denied access for whatever reasons). Does this imply there is no determinate number of marbles in the urn just because we can't know it?

 

The number of marbles in the urn is a question of metaphysics (reality). Whether we can know the number is a question of epistemology (our beliefs and knowledge).

 

The two mustn't be confused.

[Strange]

Say, there is a certain number of marbles in an urn. We do not, or cannot, know what the answer is (we're denied access for whatever reasons). Does this imply there is no determinate number of marbles in the urn just because we can't know it?

 

So one way of producing an answer is to estimate the volume of the urn - perhaps using a series of approximations about the radius at different heights - and then estimate the volume taken by each marble - again using a model of how they might be packed. If the model/theory is good it will give you a good approximation of the answer (to stick with your metaphor, you may only be able to test this by making a similar looking urn and filling it with marbles - if you don't have access to the actual number). A bad model/theory will give you an inaccurate result (e.g the model says 42 but when you build your replica it holds 142). A "wrong" theory will give an answer like -1 or "potato".

 

So, yes, there are things we can observe and we use those observations to build models (size of urn and marble) that can, hopefully, tell us things about the universe we cannot observe directly (number of marbles in the urn). But where we don't have access to the underlying reality, those models may or may not describe what is really happening - however accurate they are.

 

The number of marbles in the urn is a question of metaphysics (reality).

 

There is another level of metaphysics, which is whether the urn and marbles have any real existence outside our observations of them. That level of truth is unknowable.

[ajb]

Science tries to avoid questions that are in principle unanswerable. If you have some theory that predict the number of marbles, but you know that the number of marbles is unobservable and has no indirect effects that you could use to infer the number of marbles, then what is the point of the prediction?

[Strange]

Science tries to avoid questions that are in principle unanswerable.

 

Good point. And, ultimately, this is why questions of realism, instrumentalism or idealism are largely irrelevant to science.

[Reg Prescott]

Well, the point of all this is -- see Einstein quote again -- a scientific realist (with regard to whatever given theory) holds that things out there are a certain way independently of our minds. He maintains we are limited in what we can know about it; we might never be able to enjoy certainty (which certain posters in the thread repeatedly confuse with "truth"; the latter is a semantic notion, the former is an epistemic one -- accepting that one cannot enjoy certainty does not preclude one from being a scientific realist) in our theories, but we can at least enjoy some degree of justification in our beliefs -- by testing and confirming our theories as best we can.

 

Given a God's-eye view we could produce theories that are true simpliciter. Denied that, we can at least try to produce theories which approximate truth.

 

 

Or in the case of Einstein's watch analogy.... things in there are a certain way....etc, etc

Edited October 2, 2015 by Reg Prescott

[ajb]

Good point. And, ultimately, this is why questions of realism, instrumentalism or idealism are largely irrelevant to science.

Indeed. This is why, in my opinion, most scientists have not thought too deeply about this and indeed change their minds regularly or are just undecided. The question is rather moot from a science perspective and any philosophy in this respect will do. The only thing we keep hold of is some form of the scientific method. That is our models/ideas etc should be tested against nature.

...by testing and confirming our theories as best we can.

I think that is the 'long and short' of science.

Edited October 2, 2015 by ajb

[Reg Prescott]

Well, tell that please to Richard Dawkins who keeps insisting that evolution is TRUE.

Well, tell that please to Richard Dawkins who keeps insisting to his Creationist bêtes noires that evolution is TRUE.

[Klaynos]

Klaynos -- "How have you tested the water for impurities? Can you prove that what the dog is drinking is 100% H2O? If you cannot then how is that statement "true"? "

 

A key question indeed, and one that has been causing all kinds of confusion throughout the thread. Take a statement such as:

 

"The number of stars is odd"

 

Presumably this statement is either true or false (ignoring unnecessary complications). Would you agree?

 

Those complications, such as how you counted them, do brown dwarfs count, what are the limitations if your optical system matter though. Do no I don't think you can say that statement is true or false. Again this is an observation. Testability is fundamental else there's no point.

[ajb]

Well, tell that please to Richard Dawkins who keeps insisting that evolution is TRUE.

Okay, he says that rather casually without explaining any deeper metaphysics and philosophy of what he means by 'true'. Based on what I have read and listened to, by 'true' he does indeed mean that the idea of evolution matches the observations very well. In that sense evolution is 'true'. The theory of evolution by natural selection gives us another layer of understanding to the Universe and in particular biology.

[Reg Prescott]

Just to add a little more to my post before the this....

 

The scientific realist holds that while we may not enjoy certainty in our theories, we can at least claim a certain justification -- we have good reasons for believing that our theories approximate a true description of reality. And he might add with a hint of pride "...which is more than you nutters can claim"

[Strange]

Well, tell that please to Richard Dawkins who keeps insisting that evolution is TRUE.

 

I wonder if he would say that to a room full of philosophers (or even biologists).

 

(Not that I have much time for Dawkins' attitudes to religion.)

 

Congratulations on discovering the Quote button though.

Just to add a little more to my post before the this....

 

The scientific realist holds that while we may not enjoy certainty in our theories, we can at least claim a certain justification -- we have good reasons for believing that our theories approximate a true description of reality. And he might add with a hint of pride "...which is more than you nutters can claim"

 

I don't who "you nutters" are supposed to be.

 

The instrumentalist and the (shhh) idealist would say that we have good reasons for believing that our theories approximate a good description of our observations. (And will quietly smirk at the naivety of the realists who claim to be rational but have an unsupported, and unsupportable, belief in some sort of untestable reality. Which is really nuts.)