critiques of science

IdleRich

IdleRich
But in my heart of hearts I know that is not an explanation, and the real explanation is that we have theories about the motion of the sun and the earth and so on which have not yet failed any experimental tests so we choose to act as if they are true until they are falsified.
No. That's just shifting the problem back one, it's still inductive reasoning to say that the theories of motion which worked yesterday will work the same way tomorrow - what reason do you have to believe that except induction?
You're in the same situation as the chickens, you have only observed the laws of physics up until now, if they are going to change tomorrow then how can you predict it?
That's why the validity of inductive reasoning is so important.
 
Been away for a few days... reading all this makes me feel a bit snowed under, the target of aggression.
I'm not trying to set myself up as the arbiter of all knowledge on these things, just trying to have a discussion. I'm just a bloke who likes reading books, yunno.....

But I'll try and reply on some of these points....


For now here are some more thoughts on induction and epistemology for now, which hopefully are of interest to Mr Tea and IdleRich, and address part of what Tryptych has to say as well:

Nobody's come up with irrefutable answers to the basic problems of how we can know things. In my opinion, the way to choose between two theories which both pass experimental tests is to go for the one that offers the best explanation of what is going on. If one of the rival theories can be experimentally falsified then we can safely choose the other one as closer to representing reality.


In choosing between two rival theories neither of which have been experimentally falsified, it makes more sense to me to have a concise theory that explains the things we observe than a different theory that is the same in all respects as the prevailing one, except it adds another unexplained complication or complications.

In the "will the sun rise?" example, the prevailing theory is "the sun will rise tomorrow because the planets will continue to orbit the sun in the way Einstein's laws predict" and the rival theory would be "everything Einsteins says is true but after time x all this will change" - without any reason given for the sudden change. This is throwing away a simple explanation and replacing it with a more complicated one with no reason given for the extra complication. The new theory is the same in all respects as the original apart from an added unexplained exceptional case.
If someone had a good explanation such as "after x billion years the sun will be a red giant and the earth will be destroyed so there will be no sunrise" then that is not an unexplained extra complication, it is possibly a good explanation and as such the theory about the sun rising ought to be modified if we find there is good evidence for the sun expanding and swallowing up the earth.
there is no induction here, nobody is saying "the sun will rise tomorrow because it rose yesterday".


In Tryptych's "jumping off a building" example, the prevailing theory is "gravity will make you fall to towards the centre of the earth" and the competing theory would be "gravity will make anyone except me fall" or "gravity will make anyone fall except for this moment" without any reason given for why I or this moment are exceptional. So again the new theory is identical to the old but with an added complication or exception to the prevailing theory without any explanation.
It is not that I believe I won't fall because I believe in induction, it is that nobody has given me a better explanation than the simple one that explains very well why I will fall.


Galileo said "the stars move around the sky in that way because the earth is going around the sun" whilst the Inquisition said "the stars move around the sun exactly as if Galileo was right but actually it's angels or god causing them to move in that way and the earth is fixed" without any reason given for this deception on the part of god or huge extra complication (the existence of an all powerful god who does this strange thing with the stars every night when he could have just set it up in the way Galileo describes and chilled out with the angels).
(this example doesn't involve induction but it's a good illustration of how simple and full explanations are more powerful than complicated holey ones).



Russell's Chicken's theory is "the farmer feeds me every day because he loves me and therefore he will continue to feed me". The rival theory is "the farmer is feeding you to fatten you up and then he will kill you and sell you for profit". This theory is also more complicated than the Chicken's own theory but the crucial difference between this example and the ones above is that it contains a good explanation for why it is more complicated, so it should at least be entertained, just as the "expanding sun" theory contains good explanation of why the sun won't rise over earth in the distant future.



So in part I suppose I am appealing to "Occam's Razor" - the idea that unnecessarily complicated explanations are less likely to be true than simple ones. Emphasis on the word "unnecessarily".


"Russell's Chicken" and many other examples show that you can't rely on "the past being like the future" but when it is different in some respect, there will be an underlying explanation for why it is different and that explanation must be part of any theory that predicts changes over time.


I'm sure there are still problems with this model of theory-selection but I think it's a small step on from Popper and I find it more convincing and satisfying than anything else I have read.



We have strayed a long way from Zhao's initial thread but it's quite interesting to think about isn't it?
 

IdleRich

IdleRich
"Been away for a few days... reading all this makes me feel a bit snowed under, the target of aggression."
Oh, I'm very sorry if you felt like that, not intentional at all.
I'm glad you replied anyway because I felt that it was left hanging somewhat.

In the "will the sun rise?" example, the prevailing theory is "the sun will rise tomorrow because the planets will continue to orbit the sun in the way Einstein's laws predict" and the rival theory would be "everything Einsteins says is true but after time x all this will change"
I take your point but isn't that exactly what the chicken would say?
What I mean is, what reason is there to think that the prevailing theory will prevail tomorrow? I think the answer is induction isn't it?
 
Some more thoughts on Tryptych's post:

I don't see why saying "a theory has failed x tests, so it should not be our best guess" is any different from saying a theory has passed x tests, so it should be our best guess" - i.e. they're both forms of induction.

As soon the theory fails ONE test, then it is falsified.
It may still be used for practical calculations in the absence of a better (in purely instrumentalist terms) theory but nobody believes the theory is true any longer. How can a theory about the real world be true if it is shown to be not true even once?

A falsified theory may be "our best guess" from an instrumentalist point of view but as soon as a prevailing theory is falsified, you would hope scientists would be falling over themselves to propose replacement theories and test them out.

In fact it is almost unheard of nowadays for a theory about physics to be falsified in the absence of a "challenging" rival theory. What happens is: someone comes up with a new theory which passes all the experimental tests performed on the old theory (otherwise you can discard it straight away) and then invents a new experimental test to choose between the two rivals. At this point one of the theories will be falsified and the other will become or remain the prevailing theory.


+_+_+_

What Wikipedia has to say about Lakatos is interesting and makes me think I may have misuderstood Popper a bit! I'll have to do some more reading. I wish my local library had any books about philosophy..... you can't buy them in the airport either :slanted:

It doesn't seem like he is supporting Kuhn but rather trying to represent Popper in a different way to how the majority had perceived/portrayed his writings in order to refute Kuhn. But let's not argue about that because I haven't read any Lakatos! I'm interested to read it but not so we can argue about who knows the most about this guy or that guy, just if it makes me understand more stuff.

As to whether Kuhn was on the right track or not, I think in some respects he was and in others we was unfair. I have never been to university or spent time among academics so I don't know the truth of what goes on, only how it is represented in the writings of scientists. So I could've been hoodwinked....
It seems to me that scientists are ready to embrace paradigm change if they are given a good reason for it - look how quantum mechanics swept through the scientific establishment and before that, special and general relativity. They were successful because they gave better explanations of observed phenomena and more accurate predictions, they were big changes to the established view but they were accepted because they came with good explanations.

I'm sure there were those who didn't go along with paradigm change (Einstein famously wasn't an easy convert to quantum mechanics, mainly due to the extremely shaky interpretations given it by its originators) but QM & relativity are both striking counterexamples to Kuhn's portrayal of the scientific establishment as set hard against change, and of new ideas basically having to wait for older established scientists to die before they could be seriously entertained.

Or am I misrepresenting/misunderstanding Kuhn here? Happy to be put straight if so...

++_+_+_+_+_+_+_+_+_+_

Non-reductive materialism (Davidson?) and other non reductive philosophies, like emergent/enactive theories that Mr. Tea makes reference too, are very much on the fringes of mainstream science

yes you're right but I feel like we've moved on to me defending what I think rather than trying to stick up for the whole scientific establishment, which I've already conceded has numerous failings when compared to the ideal.
I'm quite enjoying the debate as it goes, do stop me if I'm too self-centred. :)

+_+_+_+_+_+_+_+_+_+_+_++

me - There's a key difference between "creating knowledge" and "giving you any knowledge" - science aims to find out things nobody already knew. Knowledge about literature for example is created when the literature is written, not when you learn about it. Learning about it at college or whatever is just moving knowledge from one person's brain to another's.

you - Huh? I don't understand - knowledge about science is created only on the very cutting edge of experimental science, not when you learn about it either. What's the difference? No one gets to practice "science" in college either, it's just learning about experiments and theories already created. You admit that literature can create knowledge also...

Yes you are correct in your first sentence. The difference between cutting edge science and writing literature is that science creates knowledge about the real world and that knowledge previously did not exist in any human mind. Writing literature is an act of creating something, which I will label "information" in the sense of a work of literature being a set of symbols wherein the order of the symbols matters. Something is created but it is not knowledge about the world, it is a new pattern made from the information (some of which is in the form of knowledge) already present in the brain of the author. No new truth about the world is discovered, although truth may be made more obvious or clear through literary interpretation.

This is not to belittle the creative process in anyway, it's deeply important in my opinion. But creating information/art/literature/music is clearly different to creating knowledge about the world.

Also I think you are being a little disingenuous here as you initally were saying science is not useful for "giving you any knowledge about art/literature" which really implied something along the lines of "you can't learn about Shakespeare from doing experiments" - which is what I was replying to in the post I quoted above.

(by "the world" I mean "what there is" or "reality")

_+_+_+_+_+_+_+_+_+_+_+_+_+


Now we get down to it... there is no such thing as "science done right", and science is precisely the "surrounding crap" - without that surrounding crap there would be no science.

I'm not sure what axe you have to grind here. Why do you claim there is no distinction between, say, somebody trying to find out truth about some physical phenomenon, and somebody trying to find a way to show their brand of deodorant is better than brand x or trying to get a Nobel prize at the expense of their research partners? There is clearly a difference.

Like I said before, we can agree to differ in how we define our terms. If you want you can say "science" includes all the the crap, I can say it doesn't.

But I can't understand how you can say there is nothing there APART from marketing, political infighting etc. If you really think that, why are you bothering to discuss Popper with me?

I seriously want to ask, are you just arguing this for the sake of a lively discussion or do you really not see the distinction?

+_+_+_+_+_+_+_+_+_+_+_+_+_+


The whole point of The Golem is that it is about fundamental lynchpin experiments and theories, exemplars of "proper science" and how, in fact, they are not, including Pasteur, the Michelson-Morley experiment etc. This is of course Kuhn's point too. Science at the cutting edge is determined not by appeals to "proper" objective science, but subjective factors.

Once again, I haven't read it, but once again, I have conceded many times that there are many examples of failure on the part of scientists to live up to the standards we expect of them, and I agree that they ought to be exposed.
You have to pay attention because dodgy individuals are out there, they are likely the ones in charge in a lot of cases too.
But this doesn't detract in any way from the value of pursuing knowledge in a scientific manner.

+_+_+_+_+_+_+_+_+_+


more later...


NOOOOOOOOOOOOOOOOO! :eek: :D
 

Mr. Tea

Let's Talk About Ceps
O
I take your point but isn't that exactly what the chicken would say?
What I mean is, what reason is there to think that the prevailing theory will prevail tomorrow? I think the answer is induction isn't it?

I'm not sure how this applies to the chicken Gedankenexperiment (yes, I did just use that word for the sheer hell of it), but with regards to things like the Sun failing to rise, you can actually make statements a bit stronger than induction about it. For example, lots of ways in which the Sun could fail to rise - by simply ceasing to exist, for example - would violate physical law, in this case conservation of mass-energy. If the Earth fell into the Sun for no reason this would violate conservation of momentum. These laws follow from a very general mathematical result (Noether's Theorem) which can be proven analytically. So in no Universe with physics we can understand - to the extent that we do understand at least some of the physics in our Universe - would the Sun simply fail to rise without good reason, e.g. its having undergone a nova and ceased to be a sun.
 

IdleRich

IdleRich
"For example, lots of ways in which the Sun could fail to rise - by simply ceasing to exist, for example - would violate physical law, in this case conservation of mass-energy. If the Earth fell into the Sun for no reason this would violate conservation of momentum. These laws follow from a very general mathematical result (Noether's Theorem) which can be proven analytically. So in no Universe with physics we can understand - to the extent that we do understand at least some of the physics in our Universe - would the Sun simply fail to rise without good reason, e.g. its having undergone a nova and ceased to be a sun"
No, you're not getting to the actual point. Suppose that yesterday, today and now the law of conservation of energy (or momentum for that matter) applies - how do you know it will apply tomorrow?
 
Idlerich - i was busy typing away.... started my previous long post before your one about induction. I wasn't trying to avoid your point, i hadn't read it yet.....

Induction is slippery and before I started trying to explain this on here I wasn't convinced myself that it wasn't "hand waving". But now I am convinced -

i think the thing is that we are saying "here is a theory that predicts X behaviour under circumstances Y" and that it makes no difference whether it is today or tomorrow or next year. the theory is true for all time unless something else happens before, during or after a certain time period that changes the circumstances.

The theory became the prevailing theory because it passed all experimental tests, gave good predictions and most importantly it came with a good explanation.

So we're not saying "it's like this today because it was like this yesterday" we are saying "it's like this today because nothing relevant to my very good explanation of why my theory is true has changed since yesterday".

The sun rises not because it rose before but because, yes, the earth is still rotating relative to the sun, which is still emitting light in this direction etc.


BUT: how do we know it won't just stop rising tomorow because for some reason the sun's gravity suddenly gets turned off and we ping off into deep space?

Answer: we don't, but the theories we have contain good explanations for what is going on, and a new set of theories that say "the old theories with all their explanations of gravity, light, relativity etc are completely correct but there is also an extra phenomenon which will cause the sun's gravity to get turned off tomorrow" is just a more complex version of the prevailing theories with no explanation given for the extra complexity.

We still don't know it won't happen and there might be a good explanation why gravity gets turned off tomorrow that we haven't thought of. We don't have any way to know that we are right about anything.
But I am arguing for adopting as our best theories those which contain the best explanations devoid of complicated unexplained phenomena.

The chicken may obect that our theory about the farmer is over-complicated but he cannot dismiss it out of hand because we have a good explanation for the new details.

All this largely comes from David Deutsch and it's got a lot to do with Occam's Razor I suppose.



EDIT: a lot of this post is cobblers, as idlerich has kindly ponted out!
 
Last edited:

Mr. Tea

Let's Talk About Ceps
No, you're not getting to the actual point. Suppose that yesterday, today and now the law of conservation of energy (or momentum for that matter) applies - how do you know it will apply tomorrow?

Yeah, point taken - I guess the only comeback I can make to that is "well that would mean mathematics and science would no longer be reliable as ways of understanding the Universe, and they always have been up until now", which again is a sort of meta-induction, "we use induction because it works".
 

IdleRich

IdleRich
"Induction is slippery"
Oh yes.

i think the thing is that we are saying "here is a theory that predicts X behaviour under circumstances Y" and that it makes no difference whether it is today or tomorrow or next year. the theory is true for all time unless something else happens before, during or after a certain time period that changes the circumstances.
How do you know that?
Isn't the problem of induction another way of saying "is it ok to say that a theory will hold for all time (unless something happens to change that)?"?

The theory became the prevailing theory because it passed all experimental tests, gave good predictions and most importantly it came with a good explanation.
So we're not saying "it's like this today because it was like this yesterday" we are saying "it's like this today because nothing relevant to my very good explanation of why my theory is true has changed since yesterday".
Sure but that's not getting to the nitty gritty.

BUT: how do we know it won't just stop rising tomorow because for some reason the sun's gravity suddenly gets turned off and we ping off into deep space?

Answer: we don't, but the theories we have contain good explanations for what is going on, and a new set of theories that say "the old theories with all their explanations of gravity, light, relativity etc are completely correct but there is also an extra phenomenon which will cause the sun's gravity to get turned off tomorrow" is just a more complex version of the prevailing theories with no explanation given for the extra complexity.

We still don't know it won't happen and there might be a good explanation why gravity gets turned off tomorrow that we haven't thought of. We don't have anyway to know that we are right about anything.
Exactly. Because we are using induction we don't know - but obviously our best guess is the one we have to act on. I think that you are effectively invoking Occam's Razor as a way to say that "we recognise that the theories we have depend on induction but they are the best that we have".

If the two chickens were arguing about why the farmer was feeding them and one (without any more knowledge than the other) said "Maybe in two days he is going to kill us" the other might reply (not unreasonably as you have done) "Well, Occam's Razor means that we should go with the idea that he will just keep on feeding us" - in their position that would be reasonable but as it turns out incorrect.
 

IdleRich

IdleRich
"well that would mean mathematics and science would no longer be reliable as ways of understanding the Universe, and they always have been up until now",
Well, yes that would be a circular argument (I think although according to Louise, Papineau argues that it isn't).
But I think that's why people ask the question in the first place - what basis do we have to think that any of our theories of knowledge are in any way a reliable way of understanding anything?
 

Immryr

Well-known member
If the two chickens were arguing about why the farmer was feeding them and one (without any more knowledge than the other) said "Maybe in two days he is going to kill us" the other might reply (not unreasonably as you have done) "Well, Occam's Razor means that we should go with the idea that he will just keep on feeding us" - in their position that would be reasonable but as it turns out incorrect.


if we were all to live by the 1st chickens theory we would live in constant fear of, well, everything.
 

IdleRich

IdleRich
"if we were all to live by the 1st chickens theory we would live in constant fear of, well, everything."
I was just saying that Edward's saying "gravity won't go backwards tomorrow" is analagous to the chickens from his example saying "the farmer will feed us tomorrow" which he used as an example of inductive reasoning.
Of course, I'm not saying that we shouldn't ever use inductive reasoning, I'm not just pointing out that we do use it and we should be aware of that.
 
I think my explanation shows that (or is intended to show that!) scientific knowledge should be based on falsification and the best explanations available, not on induction.

The difference between the chicken's friend and Immyr's purple gas theory is that the theory about the farmer fattening up chickens for the slaughter comes with a good explanation.
Occam's Razor in this case doesn't allow us to cut away all rival theories and adopt the simplest. But it allows us to discard those theories which are almost identical to the prevailing theory but with added unexplained exceptions.

=-=-=-=-=-=-=-=-=-=-=-

The problem of induction:
just because an experiment gave a result 100 times in a row, we have no right to assume it will be the same on the 101st attempt. so we should not use this method to create theories.



the theory is true for all time
How do you know that?

OK I don't, I was being too casual in my argument. I am still thinking this through for myself.

We don't know anything is true, you can't prove a positive. The thing is, you have a theory of how things are, it's your best theory, you might as well assume it's true and use it for predictions. It is true: There is no logical justification for doing this available to us.


Here is how theories get to be popular:
Someone invents a theory.
People argue about whether the theory's explanations seem likely.
If the explanations seem good then experiments are done to try and falsify them. (nobody bothers trying to experimentally falsify theories like "if you eat 100 doughnuts you will be able to turn invisible" because there isn't a good explanation of why this might be so to go along with the theory)
If they have a good explanation and they do not fail experimental tests then they become the prevailing theory.
In any given area of science there is a prevailing theory and then a new one comes along and one of them can be either falsified by experiment or else shown to be merely a restatement of the original theory with extra unexplained bits and therefore (I propose) less worthy of consideration as likely truth.

We have no other way to get a better theory than this - keep coming up with new theories and test them against the evidence and the explanation of the prevailing theory.

So we arrive at our best theory to date by this means, we might as well use it.

We don't KNOW if it'll be true tomorrow but it's the best thing we've got.
If we thought we knew that we would be using induction.
But we do know it's the best theory anyone's come up with so far so why not use it? It's demonstrably better than any other theory that is known to us so far.

The point is that we didn't ARRIVE at the best (to date) theory by a process of induction, we got there through falsifying rival theories or discarding them becauase of their untenable explanations.

So there is no "induction problem" in the creation of knowledge in this way.
 
Last edited:

dHarry

Well-known member
Russell's chickens are a red herring in the context of science and physics, because their situation is socio-political and psychological ("we are in the farmer's care and we trust him"). The farmer deciding to slaughter them tomorrow is a different concept to the sun/earth obeying the laws of physics (or not).

Here's a full set of lectures on the overall subject (philosophy of science, not chickens):
http://www.soc.iastate.edu/Sapp/phil_sci_lecture00.html
 
Last edited:
Yeah, my post at 3:59 was basically just going a step deeper into things but not actually answering your point, you are right!
Hopefully the one above is a bit better....:)
 

Mr. Tea

Let's Talk About Ceps
Russell's chickens are a red herring in the context of science and physics, because their situation is socio-political and psychological ("we are in the farmer's care and we trust him"). The farmer deciding to slaughter them tomorrow is a different concept to the sun/earth obeying the laws of physics (or not).

Sure, that's what I was saying, but Rich reckons the two situations are analogous (I think), in that there's no way the chickens can know that the farmer won't "arbitrarily" (from their point of view) slaughter them one day and we can't be sure - in a really logically watertight way - that the laws of physics won't arbitrarily change one day.

Have I got that more or less right, Rich?
 
Top