authority

[mooeypoo]

We already answered that question. You just never accept it, lemur.

[Edtharan]

The whole theory of scientific revolution maintains that while some of the scientific resistance to novel and challenging experimental results is just empirical scrupulousness, some is also just plain stubborn or defensive.

Yes, although there might be individuals that resist change, the scientific community as a whole is very accepting of change, so long as it is show to match reality.

 

If an explanatory paradigm is suddenly and dramatically changed by some result, all the established figures in the field become neophytes like everyone else, since they are all now alike in having to learn things from the beginning again. When you are chairman of the department and pulling in $300k a year for your absolute mastery of some theory which turns out to be wrong, not only are you exposed as dumb for not having seen it, but your hold on your prestige and money becomes tenuous, and you have to start scrambling to understand a new theory when you thought you could just coast into retirement on what you did 20 years ago. That is all so unpleasant that there is a strong motivation to hunt for something wrong in the defiant experiment rather than admit its validity, even when, in strict science, you should.

Think of it like this:

Imagine you ran a car company and are really good at it, then you design a car that doesn't work, yet you push on into production despite the evidence that it doesn't work.

 

However, other companies see that the car you design doesn't work and design cars that do work.

 

Which company will end up selling cars?

 

Actually, think about it a bit more. If you got to be the head of that car company and were successful at it, then that means you know how to choose between cars that work and cars that don't. You would have demonstrated enough competence not to make such bad decisions. Also, the head of a car company has an interest in making cars that work, not in just making cars.

 

As it is with scientists. What draws a scientist to become a scientist is curiosity and the discovery of how the universe works. You seem to think it is money, but very few scientists make a lot of money (except on TV and moves, and we all know TV is real don't we... [/sarcasm] ). They don't go in for the money.

 

So, if a scientist who became a scientist because they were curious about how the universe works, was faced with learning more about how the universe works, or a small amount of money, which do you think they would take?

 

This is why your analogy is so wrong. Sure, there might be a few scientists out there that would act as you stated, but that is only a few out of millions. And, if scientists acted that way, then they would loose their job (and there have been examples of such).

 

In other words, if the chairman in your example did that, because everyone else knows that they were wrong, and the chairman continues to act as if they weren't, they would have an even more tenuous hold on their job and would likely be fired for incompetence.

 

It is often not just that the experimental result is doubted beyond all reasonable grounds for skepticism, but that it is worked into the existing paradigm via contrived elaborations designed to preserve the paradigm at any cost. If you look at the cycles and epicycles that eventually adorned the Ptolemaic theory of solar system, or the artificial solutions offered to explain away the Michaelson-Morley experiment while still preserving the old aether hypothesis, you wonder how people could have been so foolish as to think these constructions made sense.

It is interesting that you use the cycles/epicycles as an example. It shows how little you know of science. It was the fall of the cycle/epicycle proposition that really founded science. Scientists didn't really exists (they were called natural philosophers then and operated under a different methodology) and it was Newton that showed how experimentation and the scientific method could work.

 

In other words, by bringing up the epicycle model, you have shown exactly why the scientific method works over the other methods. You have provided a really good argument against your position. You are doing our work for us.

 

What is often required is simply a new look at things that brings the old paradigm crashing down, since the experiment itself often fails to speak clearly to its implications. Thus when Galileo dropped two balls of different weights from the leaning tower of Pisa and noted that the timing of their striking the ground failed to correspond to the result predicted by the reigning Aristotelian physics, he had to admit that there still was a difference in the time the balls struck, as the Aristotelians maintained, but that it just wasn't as large as it should have been. Does this result prove anything? It depends on whether you discount the difference as due to an undetected extraneous factor or not, and the experiment doesn't tell you that, but only your own sense of what the small apparent difference must mean.

Actually, Galileo probably never performed that experiment. What he did show was that the reasoning behind the old model was flawed. Through showing that if you had a lighter weight that fell slower than a heavy weight, then joining them together should make them both slower (as the lighter weight would slow down the heavy weight) and faster (the weigh of them combined would be greater than separate and thus must fall faster than the initial heavy weight) at the same time.

 

At that point, one had no evidence for either, it was only though very careful experimentation was one discovered to be correct.

 

He also showed that different shaped weights, even though the same weight, landed at different times. His actual experiments used weights the same size and shape, but of different weights (balls rolled down slopes) to determine the difference (none between different weighted objects).

 

But here the thing. Even though we accept Galileo's theory, we don't stop testing them. We use vacuum pumps to create airless environments to test it, we have done the experiments on the moon where there is a natural vacuum, and we use highly sensitive measuring devices (far more accurate than Galileo could have ever imagined or though possible). In other words, we still have not 100% accepted Galileo as being right. Any result, no matter by who, or how long it has been accepted, is still under scrutiny by science.

 

And that is the power of science. It is why it works.

[Marat]

Ptolemaic astronomy, just like modern astronomy, was based on empirical observations! It was just using the wrong orientation point to characterize the relative motions observed. The change from Ptolemaic astronomy to Copernican astronomy can be found in every history of science text: Ptolemaic astronomy was not simply magic, such as, for example, Renaissance medicine was in its doctrine of cures by similars, astrological calculations, precious gems, etc.

 

I think the reason why professional reputation and careerist calculations which are properly external to science can still have a major influence in the acceptance or rejection of new scientific results and new theoretical paradigms is that experiments are seldom entirely clear and decisive. Consider for example Galileo's discussion of the demonstration of 'Galilean relativity' of motion in his example of a knife dropped from the mast of a moving ship, which still lands as the base of the mast rather than behind it. The contemporary Aristotelians, whose doctrine was still being taught at universities into the mid-18th century, could argue against it that the result Galileo observed was too small to refute their position. The case was even less clear for Copernicus, who lacked the physics at the time of his theory (which he presented only as a hypothetical) to provide a physical basis for it. According to the physics he was using, his theory of the solar system, if true, would have meant that the birds would fall out of the trees as the Earth moved around the Sun, given that he was writing ca. 120 years before Galilean relativity. This would give plenty of room for the defenders of the Ptolemaic theory to refuse to accept Copernicus' account and still seem to have good scientific theory on their side.

 

Or even look at the situation today in diabetology. Duncan Adams has proved that changes in the retinal pericytes of diabetics can only have been caused by autoimmune processes and not by hyperglycemia, and yet this is so far off the mainstream, which tries to explain all the vascular and neurological changes in diabetes by hyperglycemia, that almost no one accepts Adams' result, and in fact most researchers in the field have never even heard of it. So Dr. Adams in New Zealand is recommending that patients be treated with immunosuppressive rather than hyperglycemia normalization, while everyone else is confined to the hyperglycemia route. Why is this result simply being neglected? Logically it should start a panic or a revolution, as Adams himself noted in his paper, but there is just general silence, as though it never happened, because it is simply too disrupting to deal with.

 

To me this looks like the start of what historians of scientific revolution call 'abnormal science,' when there is a clearly stated challenge to the existing explanatory paradigm but no one has yet tried to deal with it. In theory, there will now ensue a period of attempts to accommodate this discomforting result, either by challenging the empirical validity of the study or trying to contrive odd devices to fit the result into the existing explanatory paradigm. Only if the existing paradigm accumulates sufficient conceptual burdens from these efforts will scientists finally decide to abandon the paradigm and adopt a new one.

 

But what this model, developed by Thomas Kuhn and extended by Imre Lakatos, suggests, is that scientists don't always behave like good scientists and just respond neutrally to each new empirical result as it emerges, but instead they borrow from their existing theories to add or subtract weight to what is reported and direct its pressurs either away from the theory or perhaps eventually permit them to operate against the theory -- but only after considerable resistance. You wouldn't say that physicists were operating with a pre-scientific natural science in the 1880s when their response to the Michaelson-Morley experiment was to twist and swirm out of it to save the aether hypothesis at all costs.

[Edtharan]

Ptolemaic astronomy, just like modern astronomy, was based on empirical observations! It was just using the wrong orientation point to characterize the relative motions observed. The change from Ptolemaic astronomy to Copernican astronomy can be found in every history of science text: Ptolemaic astronomy was not simply magic, such as, for example, Renaissance medicine was in its doctrine of cures by similars, astrological calculations, precious gems, etc.

I don't really get what you are trying to say here. To me it sounds like you are trying to say that before the scientific method was formalised, they didn't follow the scientific method (but something similar).

 

If that is what you are saying, then you have no objection here. The method they followed was similar, but it didn't have the "falsifiability" requirement that the modern scientific method has. And, because of that the problems you point out with it would occur. No surprise or objection ehre.

 

The modern Scientific Method is different from the ancient methodologies. Now, while they were somewhat successful, they had many problems, one of which was a reliance on authority. That is if someone was considered an authority, then they must be true even if there is evidence against them.

 

You seem to think the modern scientific method is like this, but that was the method used by the the pre scientific investigators. It was through the repeated failure of this methodology (even though it did have some successes) that the modern scientific method was created. It was because there were more and more things being shown not to be true, even though the authorities that initially claimed them to be true, that they recognised the problem and sought to work out a better system (the modern scientific method).

 

They even changed the name of the field to indicate that the approach was fundamentally different (science as opposed to natural philosophy).

 

I think the reason why professional reputation and careerist calculations which are properly external to science can still have a major influence in the acceptance or rejection of new scientific results and new theoretical paradigms is that experiments are seldom entirely clear and decisive. Consider for example Galileo's discussion of the demonstration of 'Galilean relativity' of motion in his example of a knife dropped from the mast of a moving ship, which still lands as the base of the mast rather than behind it. The contemporary Aristotelians, whose doctrine was still being taught at universities into the mid-18th century, could argue against it that the result Galileo observed was too small to refute their position. The case was even less clear for Copernicus, who lacked the physics at the time of his theory (which he presented only as a hypothetical) to provide a physical basis for it. According to the physics he was using, his theory of the solar system, if true, would have meant that the birds would fall out of the trees as the Earth moved around the Sun, given that he was writing ca. 120 years before Galilean relativity. This would give plenty of room for the defenders of the Ptolemaic theory to refuse to accept Copernicus' account and still seem to have good scientific theory on their side.

Again, this is all pre-scientific method, so has no real relevance to the discussion other than to point out that the old system had its flaws and the modern scientific method was created to fix them.

 

In other words, you are making my argument for me. Thanks.

 

Or even look at the situation today in diabetology. Duncan Adams has proved that changes in the retinal pericytes of diabetics can only have been caused by autoimmune processes and not by hyperglycemia, and yet this is so far off the mainstream, which tries to explain all the vascular and neurological changes in diabetes by hyperglycemia, that almost no one accepts Adams' result, and in fact most researchers in the field have never even heard of it. So Dr. Adams in New Zealand is recommending that patients be treated with immunosuppressive rather than hyperglycemia normalization, while everyone else is confined to the hyperglycemia route. Why is this result simply being neglected? Logically it should start a panic or a revolution, as Adams himself noted in his paper, but there is just general silence, as though it never happened, because it is simply too disrupting to deal with.

Actually, it has been known for quite some time that certain types of diabetes is caused by autoimmune responses. So someone coming out and making this "new" claim would not really make too much of an impact.

 

it is like if I came up to you and said: "Did you know the sky is blue, I can prove it." You probably would not be surprised that it is blue because you would have already know that.

 

The reason the research is being done on hyperglycaemia (or hypoglycaemia) for diabetes is that is the effect of it that causes the symptoms and damage to the body. Knowing it is an autoimmune condition (type 1 diabetes) still causes the problem of poor control of blood glucose levels. Type 2 diabetes is caused by stressing the pancreas with too much glucose in the blood stream (actually it is more about going from low to high blood glucose level more than just constantly high level, but constantly high level also stresses the pancreas).

 

However, in either type 1 or type 2 diabetes, the fluctuating blood glucose levels puts further stress on the pancreas exasperating the problem. If medication and medical treatments can be found to reduce this stress on the pancreas, then the problem can be stabilised, and possibly reversed.

 

However, this does not mean that research into eliminating type 1 diabetes through dealing with immune system is not being undertaken. There is much research being done with transplants, artificial pancreas, tissue engineering to replace a pancreas and re-educating the immune system to eliminate the autoimmune response.

 

To me this looks like the start of what historians of scientific revolution call 'abnormal science,' when there is a clearly stated challenge to the existing explanatory paradigm but no one has yet tried to deal with it. In theory, there will now ensue a period of attempts to accommodate this discomforting result, either by challenging the empirical validity of the study or trying to contrive odd devices to fit the result into the existing explanatory paradigm. Only if the existing paradigm accumulates sufficient conceptual burdens from these efforts will scientists finally decide to abandon the paradigm and adopt a new one.

As I said above: This study didn't actually tell them anything they didn't already know, or were already performing research on to correct.

 

It also means that this "evidence" for your argument that science rejects things is clearly false, because they already knew this, accepted this and were working on a way to treat it before this study was even begun.

 

But what this model, developed by Thomas Kuhn and extended by Imre Lakatos, suggests, is that scientists don't always behave like good scientists and just respond neutrally to each new empirical result as it emerges, but instead they borrow from their existing theories to add or subtract weight to what is reported and direct its pressurs either away from the theory or perhaps eventually permit them to operate against the theory -- but only after considerable resistance. You wouldn't say that physicists were operating with a pre-scientific natural science in the 1880s when their response to the Michaelson-Morley experiment was to twist and swirm out of it to save the aether hypothesis at all costs.

Yes, it is important to defend the old theories. This could be because the results of the new experiments are wrong, that the old theory does indeed explain the new data just that no one was aware of that application of it before, or for many other reasons.

 

defending an old theory does not mean that that the scientific method has failed, it actually means it is working. For a scientist to defend an old theory over a new theory means that the scientist is seriously considering that the new theory is on equal terms as the old theory. It means that they are not just accepting of authority (and someone saying: "I have new data" is an authority on the new data).

 

The purpose of the scientific method is to determine which of competing explanations of a phenomena is the better one. And the only way to do this is for someone to defend each theory and try their bets to prove the others are inferior.

 

You seem to be saying on one hand that scientists will never accept anything new. But then on the other hand you seem to be saying that scientist should not just accept anything new because it is new.

 

This is a logical fallacy called a strawman. You have created your own version of what the scientific method should be, and does so in a way that is completely impossible to achieve. then when science does not achieve it, you claim victory.

 

What we have had to repeatedly explain to you, and you don't seem to be able to grasp, is that your version of the scientific method is wrong. So you can argue all you like against your own made up version of it, but your arguments don't have any relevance to the actual scientific method.

 

We haven't even been able to disuses the scientific method with you, because you can't even seem to understand that you don't know what the scientific method actually is.

 

For this discussion to proceed, you have to understand what it is you are arguing against. You must forget your version of the scientific method, and learn what the scientific method actually it.

[mississippichem]

Its worth mentioning that last year, two chemists from China fabricated some x-ray diffraction crystal structures. They took an established compound from the literature and basically superimposed different atoms on the crystal structure in order to claim they had a novel crystal structure. Their results were published in a Wiley-Interscience journal. They got away with it for a while until someone noticed that their calculated unit angles were not allowed by crystal packing rules. The two chemists' careers were ruined, they were banned from publishing in many journals and I believe they lost what grant money they had IIRC.

 

We have zero tolerance for scientific fraud in the professional world. If you fabricate data, you will be found out. There is no one scientist smart enough to fool all of their peers in the field. Eventually someone comes along with a nose for BS and the fraud perpetrators get sacked.

 

Experiments get reproduced all the time. Lets say someone reports a new compound. I might be interested in making a new derivative of that compound. So first I try their reported procedure out to make their compound, before I try my new alterations out. If their procedure doesn't work, I will assume it is my fault and seek the consultation of a more experienced chemist. If he can't make it work, then perhaps we've found a mistake [or a rat] so we report to the journal and perhaps even publish a correction. This stuff happens on a daily basis.

[Marat]

It's a huge question to try to say when the modern scientific method actually came into existence. Bacon made a major contribution with his emphasis on empirical study, and Descartes made an important addition with his concentration on quantification and skepticism, but there is no one point where you can say that here we have a period clearly prior to the modern scientific method while here we are clearly within it. Elements of the modern scientific method were present or absent in various scientists at various times. Thus Hippocrates was already insisting on careful matching of hypotheses to observation, and in Ancient Egyptian medicine you can find excellent clinical descriptions right next to pure magic. Tycho Brahe was doing meticulous observational astronomy while operating under an outmoded hypothesis. His assistant, Kepler, was modern enough to insist that the greater accuracy of Brahe's data be more closely represented by a better hypothesis, but old-fashioned enough to be reluctant to accept that anything so 'imperfect' as an ellipse represent the planetary motions rather than a circle.

 

The falsifiability criterion for testing scientific theories is an invention of the mid-20th century philosopher of science, Sir Karl Popper, and was not identified as a methodological principle of science before then. It was even highly controversial for the first few decades after Popper stated it, though now it is coming into general acceptance as accurately describing what science does or should do.

 

The use of the term 'natural science' rather than 'natural philosophy' was not in response to any sudden discovery in the history of science that the previous methods had been wrong and the new ones were right; after all, even Newton conceived of himself as doing natural philosophy. The full title of the 'Principia,' after all, is 'Philosophiae Naturalis Principia Mathematica,' or 'The Mathematical Principles of Natural Philosophy.'

 

I am not sure this discussion is really so much about the scientific method, since the whole theory of scientific revolutions as developed by Kuhn and his followers since the 1970s really doesn't focus on that, but rather, on how scientists using a valid scientific method respond to data which contradicts established theories. There is an essential tension here, since, as I mentioned before, normal science has two responses to data which don't fit the theory. One response is to reject extremely challenging new data as simply 'having to be wrong' because it doesn't fit the established paradigm of explanation. This is the response of scientists to reports of ghosts, levitations, perpetual motion machines, etc. Because they are off the scale, they are not even allowed to be true, and even if serious defenses are made of these things with measurements, observations, and experiments, the initial response is to try to feed the conceptual pressure which results generate away from the reigning hypothesis, which has to be preserved at all costs, and to explain them away as experimental errors, extraneous variables, inaccurate statistical methods, poor smoothing out of the curve, etc. Even if these results are admitted, their full revolutionary implications for the established theory are rejected, and they are built into the existing theory via ad hoc constructions which preserve the theory by assigning some sort of exceptional or non-threatening status to the recalcitrant data. But as these accommodations and their accompanying conceptual tensions accumulate, the theory starts to look so overburdened by these ad hoc devices (e.g., epicycles) that it is eventually but only reluctantly abandoned. Black was still using the material theory of heat at the beginning of the 19th century even though Rumford's result was decades earlier; phlogiston theories were still around a generation after Lavoisier's work; and in the 1926 edition of the 'Yale Medical Journal,' there was a paper arguing for digestive disorders as the true cause of diabetes, since this had been a leading theory before the work of Banting and Best in 1921-22, which clearly proved that a failure of the pancreas to produce insulin was the cause of type 1 diabetes.

 

It is thus in the institutional reluctance to match hypotheses to startling new data that the unscientific nature of science emerges, rather than in any actual problem with the use of the details of scientific method. Even Aristotle thought that his theory should match the data, so the reluctance of Einstein to accept the implications of quantum mechanics for causality, and of modern scientists generally to accept the Michaelson-Morley result, the wave-particle duality of light, etc. is really not an issue of modern versus antique scientific method.

 

I guess I explained badly my example of the reluctance of modern medical science to accept the new data challenging the hyperglycemia hypothesis of complications. No one doubts that type 1 diabetes is caused by autoimmunity, since that has been increasingly accepted even since the 1960s. Interestingly, the cause of type 2 diabetes is also now coming to be seen to involve some autoimmune factors, so the two disease subtypes may be more similar than once thought.

 

But the problem now is with what causes the vascular and neurological complications as late sequelae of the diabetic state after the initial disease has already come into existence through genetics, some unknown environmental trigger, and autoimmunity (type 1), or genetics, perhaps some autoimmune effects, and lifestyle triggers (type 2). The existing theory has been that the sole cause of complications in diabetes is the hyperglycemia which cannot be safely corrected by insulin or the various oral medications now available. But what Duncan Adams' research shows is that the same autoimmunity which caused the patient's diabetes to come into existence in the first place by destroying the pancreatic beta cells which produce insulin also persists throughout the life of the patient and goes on to destroy other tissues as well, such as the retinas, the kidneys, the cardiovascular system, and the nervous system -- rather than hyperglycemia being the agent of this damage. (See Duncan Adams, "Autoimmune Destruction of Pericytes as the Cause of Diabetic Retinopathy," Clinical Ophthalmology, 2 (2) 295 (2008). If Adams is right, then the entire approach to diabetes treatment is radically wrong, yet his results are now three years old and no one wants to discuss them. No matter how good the underlying scientific method is, if the scientists simply look away from results which are too challenging to the reigning theories, then the method is all for naught.

 

I guess

[Edtharan]

I am not sure this discussion is really so much about the scientific method, since the whole theory of scientific revolutions as developed by Kuhn and his followers since the 1970s really doesn't focus on that, but rather, on how scientists using a valid scientific method respond to data which contradicts established theories. There is an essential tension here, since, as I mentioned before, normal science has two responses to data which don't fit the theory. One response is to reject extremely challenging new data as simply 'having to be wrong' because it doesn't fit the established paradigm of explanation.

If you re read my post, you will see I actually addressed this. The reason that there has to be resistance, even in the face of new data is: that that data might be wrong.

 

This means that if someone claims new data that contradicts accepted theories, then that data had better not be the result of an error. The longer a theory has been accepted for, the more data supporting it there is. But if someone then produces new data that contradicts it, then there needs to be a lot of resistance because it is more likely to be an error (as it would have shown up earlier if really it was there).

 

But yes, sometimes the new data is correct and not the result of an error (or forgery), and when this is confirmed, then the new data gets accepted.

 

There is not one theory in science that has not been changed due the the dicovery of new data that contradicted the other theory. Not one. This proves that science is willing to accept new data because those theories would never have changed if they didn't, and since they have, that is proof.

 

This is the response of scientists to reports of ghosts, levitations, perpetual motion machines, etc. Because they are off the scale, they are not even allowed to be true, and even if serious defenses are made of these things with measurements, observations, and experiments, the initial response is to try to feed the conceptual pressure which results generate away from the reigning hypothesis, which has to be preserved at all costs, and to explain them away as experimental errors, extraneous variables, inaccurate statistical methods, poor smoothing out of the curve, etc.

This is the whole thing I have been talking about in this post and my last post. Science dosn't just accept new data because it is new data. The new data has to be checked for errors, mistakes and fakery. It is only when these checks are done (which you call rejection because you only point to the instances when they were found to be such - that is a logical fallacy called Cherry Picking: http://en.wikipedia.org/wiki/Cherry_picking_(fallacy) ) and the data found to be reliable can the data be used as an argument against a theory.

 

The data supporting the theory also went through this process, so it is only fair that any other data (supporting or opposing) also goes through this same process and requirement.

 

Just because it is new data, does not automatically make it good data.

 

Even if these results are admitted, their full revolutionary implications for the established theory are rejected, and they are built into the existing theory via ad hoc constructions which preserve the theory by assigning some sort of exceptional or non-threatening status to the recalcitrant data.

Yes, this is right. If the current theory only need a small change to account for the new data, then you may as well use the modified theory. This is not a bad thing, it is a good thing. Just because a new theory to explain the data is proposed, does not make it a good one.

 

Take for example the theory that describes how an aeroplane wing works. There are at least 2 of them. Both give the same results, but each works better given how you are looking at the situation.

 

Theories are descrioptions, and only that. They are descriptions of the regularities that we observe in the universe. they are not the actual "source code" (so to speak) of the universe. This means that there can be several equally valid ways to describe a particular phenomena. However, some work better in certain situations than others.

 

The principal of Occam's Razor is the guiding principal here. If a theory is too complex, even if it gives the same result, then there is no reason to use it if a simpler theory works just as well (occasionally, the situation determines which is simpler as it is with wings).

 

But as these accommodations and their accompanying conceptual tensions accumulate, the theory starts to look so overburdened by these ad hoc devices (e.g., epicycles) that it is eventually but only reluctantly abandoned.

Yes, this is actually a good thing. Remember, theories are just description (it is why people keep telling you to learn what a scientific theory actually is, rather than use your own definition). It is better to use the simplest description you have for a given phenomena because it means there is less for you to make a mistake on. So as data build up about a phenomena, the description of it tends to become more complex. then when someone realises there is a simpler way to describe it, then the new simper description (theory) gets used.

 

With epicycles, the description was based on circles, but as the data kept coming in, the theory (the number of cycles) kept growing. Now, there is mathematics called Fourier Analysis (http://en.wikipedia.org/wiki/Fourier_analysis) which deals with how complex shapes can be built up form simple shapes. This means it would be possible to construct the orbits of the planets accurately using epicycles.

 

In other words, the epicycle theory still works. It is a perfectly valid and operational theory. However, you would actually need an infinite number of such circles to accurately predict the orbits. That is a problem.

 

However, if one uses ellipses rather than circles, then the mathematics of it becomes much simpler.

 

So, why would you embark on an infinite number of calculations, when you could just do a couple of simple sums to work out the orbits of a planet.

 

And that is why the epicycles theory is not used any more.

 

There are 2 reasons an hypothesis gets rejected:

1) It doesn't fit the data.

2) There is a simpler description that accounts for the data.

 

there have even been occurrences where a simpler description was though to be found, only later that data was discovered through experimentation that that showed that the theory was too simple and couldn't account for the data.

 

Light is a really good example as this occurred several times (between particles and waves). What it was, was that there was a simpler explanation that accounted for all the data, it was just that it hadn't been thought of at the time.

 

Black was still using the material theory of heat at the beginning of the 19th century even though Rumford's result was decades earlier; phlogiston theories were still around a generation after Lavoisier's work;

Yes, as that description of the phenomena was accurate enough for his work, and simple enough. However, the phlogiston theory does not account for all the data, but that discrepancy didn't effect the result of the work.

 

and in the 1926 edition of the 'Yale Medical Journal,' there was a paper arguing for digestive disorders as the true cause of diabetes, since this had been a leading theory before the work of Banting and Best in 1921-22, which clearly proved that a failure of the pancreas to produce insulin was the cause of type 1 diabetes.

A theory does not change over night. It can take decades to confirm or invalidate data, especially in medicine where the course of a disease might take years to develop (which means any study done on the development of the disease also takes years to perform).

 

There is only a 4 to 5 year difference there, and such studies take decades to do. So I am not surprised that there are still competing descriptions occurring at that time.

 

In fact, if what you are claiming about science is correct, then you would never get such a dispute occurring. The fact that it does occur is proof that your claims are wrong.

 

It is thus in the institutional reluctance to match hypotheses to startling new data that the unscientific nature of science emerges, rather than in any actual problem with the use of the details of scientific method.

Again, you havae what science is completely wrong, and you are only argueing against your incorrect concept of science.

 

Stop. Listen.

 

Yopu do not ahvae the correct view of science. You need to readjust your undersntaing first. All you are doing here is to continue, against the evidence counter to it, to retain you own description of what occurs in scinece.

 

Guess what: This is exactly what you are claiming sicence does.

 

Perhaps the problem is not science it is your attitudes to new information you are projecting onto science.

 

Even Aristotle thought that his theory should match the data, so the reluctance of Einstein to accept the implications of quantum mechanics for causality, and of modern scientists generally to accept the Michaelson-Morley result, the wave-particle duality of light, etc. is really not an issue of modern versus antique scientific method.

It is not individuals that we are talking about. I have already agreed that individuals will do this. But you keep using individuals and the scientific community interchangeably. This is not a correct view, it is in fact a logical fallacy: Fallacy of Division ( http://en.wikipedia.org/wiki/Fallacy_of_division ).

 

Just because some individuals might not be able to accept change does not mean the scientific community as a whole is resistant to changes.

 

I guess I explained badly my example of the reluctance of modern medical science to accept the new data challenging the hyperglycemia hypothesis of complications. No one doubts that type 1 diabetes is caused by autoimmunity, since that has been increasingly accepted even since the 1960s. Interestingly, the cause of type 2 diabetes is also now coming to be seen to involve some autoimmune factors, so the two disease subtypes may be more similar than once thought.

Diabetes is a complex disease. It has many factors that can cause it (including genetics). However, the "disease" is really a list of symptoms, not a cause. Thus there will be many causes of a set of symptoms.

 

The problem is not that you described it badly, but that you didn't actually understand what you were trying to describe and thus your argument from it was not a good argument. this is the exact same problem you are having with science and the scientific method.

 

Until you stop and actually try to understand what it is you are arguing about, you can not form a coherent argument. This is not the speculation sub-forum, and as such there is a certain level of understanding needed to be able to discuss a topic.

 

If you don't understand the topics under discussion, either take the time to learn about them (instead of just repeated making the same mistakes with them) or ask to get the topic moved to Speculations.

 

But the problem now is with what causes the vascular and neurological complications as late sequelae of the diabetic state after the initial disease has already come into existence through genetics, some unknown environmental trigger, and autoimmunity (type 1), or genetics, perhaps some autoimmune effects, and lifestyle triggers (type 2). The existing theory has been that the sole cause of complications in diabetes is the hyperglycemia which cannot be safely corrected by insulin or the various oral medications now available.

Insulin is not a treatment to correct diabetes, it is a treatment to manage the symptoms of the disease. Again, you are showing you don't actually understand enough about diabeties to for an argument based on it. Either learn about it, or don't use it.

 

But what Duncan Adams' research shows is that the same autoimmunity which caused the patient's diabetes to come into existence in the first place by destroying the pancreatic beta cells which produce insulin also persists throughout the life of the patient and goes on to destroy other tissues as well, such as the retinas, the kidneys, the cardiovascular system, and the nervous system -- rather than hyperglycemia being the agent of this damage. (See Duncan Adams, "Autoimmune Destruction of Pericytes as the Cause of Diabetic Retinopathy," Clinical Ophthalmology, 2 (2) 295 (2008). If Adams is right, then the entire approach to diabetes treatment is radically wrong, yet his results are now three years old and no one wants to discuss them. No matter how good the underlying scientific method is, if the scientists simply look away from results which are too challenging to the reigning theories, then the method is all for naught.

 

I guess

As you said "if" Adams is right. In other words, the results of Adam's study is not enough to prove him right. However, as I said, there is research already being done to correct the autoimmune aspect of diabetes. But you are saying that because of Adam's study, they should start researching to correct the autoimmune aspect of diabetes.

 

This is why there has not been a major shake up of the medical scientific community over this study, because it if it is true, they should just keep doing what they already have been doing. Yes, keeping the status quo is a real revolution.

 

the thing is, treating the hyperglycaemia is still necessary. Because of the loss of function of the pancreas the people with diabetes still end up with low blood sugar and still pass out from low blood sugar and all the other problems of it. Sure, the progression might be caused by autoimmune problems if Adam's is correct, but you still have to deal with the rest of the symptoms of the disease.

 

As I said above, with medical research it can take decades for data to be confirmed and tested for validity. Jsut because 1 person produced some result does not mean that theory changes overnight. Adam might be wrong, but it could take 5 years to confirm this. Medical research on long term diseases take time. Just because in the movies they find a disease, come up with a theory to its cause and then create a treatment to fix it in 3 hours, does not mean that real life medical research occurs on the same (or even similar) time scales. It generally takes about 10 years for a treatment, once identified in a lab to make it to human trials (it has to go through testing to be sure it does what it is supposed to do, animal trials to make sure there is not any major side effects that might occur, human trials to make sure it work in humans and doesn't have any side effect specific to humans, etc).

[Marat]

I think with respect to the diabetes sub-topic there is still some confusion about what I am saying. I am not trying to discuss diabetes per se, considered (in type 1) as a pancreatic insufficiency in which autoimmune processes have destroyed the beta cells so little or no insulin is produced, and for this condition exogenous insulin is required. That's the problem from 1921, and insulin treatment ever since then has had its attendant problems in insufficient normalization of hyperglycemia and causing the unwanted and often lethal side-effect of hypoglycemia. These problems, as you quite rightly state, would to some degree remain even if the late sequelae of the disease, the microvascular and neurological complications, were successfully treated or not by addressing the continuing autoimmunity, which Adams and a few others are now starting to say not only causes the initial beta cell destruction, but also goes on to cause the vascular and neurological lesions which start to appear around a decade after the beta cell destruction.

 

But the key point is this. Much of the problem with insulin therapy now is that it is directed toward the scrupulous normalization of erratic blood glucose levels, and this is not only difficult, imperfect, and dangerous, but it also seems to be not very effective at preventing the vascular and neurological late sequelae of the disease. But if it could be shown, as Adams and others are now doing, that the vascular and neurological late sequelae are in fact not really due to the toxicity of hyperglycemia at all, but are really due to the continuing autoimmune disease, then this would be revolutionary. The present demand for strict blood sugar control could be abandoned; blood sugar control could be relaxed so as just to avoid ketoacidosis, maintain weight and energy, and avoid hyperosmolar coma, but not to court the lethal hypoglycemia which now accounts for 4% to 6% of all type 1 diabetic deaths. Instead the focus could be on using immunosuppressive drugs to treat the continuing autoimmunity so as to interrupt the processes which are actually causing the vascular and neurological complications. This is in fact what Adams recommends.

 

Thus I of course agree with your statement that insulin is not a treatment to correct diabetes but a treatment to manage the symptoms of the disease, at least if we take hyperglycemia as its presenting and most prominent symptom. But if you notice, you presented that comment as a correction of my statement that "the existing theory has been that the sole cause of the complications of diabetes is the HYPERGLYCEMIA which cannot be safely corrected by insulin... ." So what I was saying cannot be safely corrected by insulin was not, as you suggest, 'diabetes,' but rather, simply 'hyperglycemia.' Type 1 diabetes cannot be corrected by anything, other than islet or whole pancreas transplant, and both of these work poorly and are rarely applied in patients who do not already require immunosuppression for other reasons.

 

I think the difference between our views on the wider issue may again just be a miscommunication. I don't dispute that new scientific data do not typically justify a total and radical transformation of the existing theories without many years of further testing. I also agree that when science does take data seriously, it should and does apply the scientific method to their evaluation. But my suggestion is that science often simply refuses to turn its mind in any serious way to data or experimental results which seem too challenging. This is why, for example, Adams' study -- even though if true it would have revolutionary significance -- is still being neglected. I agree that it needs further study and independent confirmation before it could require serious changes in the existing paradigms of diabetology, but my point is rather that it is not attracting those studies, and that there is no good scientific reason why it is not. Instead, the existing explanatory paradigm appears to be protecting itself from this unsettling result by simply refusing to enquire further into it and carrying on as though it had never been published in a peer-reviewed journal. This attitude stands outside of scientific method, as I argued earlier, since it is rather a refusal to engage with the recalcitrant findings.

 

Refractory findings will always generate conceptual pressure, but whether that pressure is fed against the existing explanatory paradigms or fed away from them does a lot to change how significant the data are allowed to become (cf. W. V. Quine, 'The Web of Belief.') This step is a conceptual decision about what to do with the data, and again is not directly part of the use of the scientific method. Thus to take a few examples from 18th century science, some chemists identified phlogiston with what we would today recognize as carbon, while others thought of it as weightless. In both of these cases, experimental results which we might want to say proved that no phlogiston was present could not count against the theory -- not because the scientific method was not being properly applied -- but because the theory structure was designed to feed these conceptual pressures away from challenging the explanatory paradigm. I.e., yes, phlogiston is there, but it's in the carbon, or yes, it is there, but it has no weight or negative weight. Similarly, the supposed electrical fluid was thought to be pumped out of the ground by the electrical accumulators of the era, so this theoretical presupposition warped the significance of any data offered in tracing how electricity interacted with other phenomena. The same trick could be pulled with heat fluid by assuming that it was invisible and had either no weight or negative weight, with the result that correct application of the scientific method within that theoretical structure could not disprove the theory. Theories in this sense act almost like biological entities, designed to preserve their own life against unhealthy environmental influences, such as refractory data in the case of scientific paradigms.

 

Nothing of what I am saying about the often less than rational self-preservation of scientific paradigms is at all new or revolutionary. Thomas Kuhn's 'Theory of Scientific Revolutions' in the 1970s already stated these views, and Imre Lakatos in the 1980s offered some more empirically developed versions of his theory of scientific revolution.

Edited June 11, 2011 by Marat

[DrRocket]

I think with respect to the diabetes sub-topic there is still some confusion about what I am saying. ...

 

Nothing of what I am saying about the often less than rational self-preservation of scientific paradigms is at all new or revolutionary. Thomas Kuhn's 'Theory of Scientific Revolutions' in the 1970s already stated these views, and Imre Lakatos in the 1980s offered some more empirically developed versions of his theory of scientific revolution.

 

Medicine is not science.

 

The objective of science is understanding of natural phenomena. The objective of medicine is treatment and cure of disease and injury, using the best available knowledge and methods.

 

Medicine makes use of science, principally chemistry and biology, but it is a sub-discipline of neither. In many regards medicine is similar to engineering, which also is a user of science, but is not science.

 

A physician treats a patient using the knowledge and techniques available to him, despite acknowledged gaps in that knowledge and limitations in techniques. He is faced with a most complex and variable system, the human body, and offers treatment despite uncertainty in outcome and often without definitive knowledge of the root cause of an ailment. The art of medicine, like the art of engineering is practiced in the face of uncertainty, within time constraints, and is limited by cost considerations. Medicine even more so than engineering is constrained by licensing and other governmental regulation. Medical procedures and treatments are subject to extensive governmental review and the unavoidable influence of politics and lobbying.

 

It is inappropriate to criticize medicine for failure to follow strict scientific discipline. It is equally inappropriate to criticize science on the basis of medical protocol.

[Ophiolite]

Marat's position appears to be that there is a degree of pressure not to disturb current paradigms, especially if the disturbance would be substantial. That is a sound and, to me, obvious position to take. An example that occurs to me is the former strong disposition against any hint of catastrophic thinking within geology. Before Alvarez proposed that an impactor caused the KT boundary extinction event two Nobel laureates had suggested similar processes. Both were ignored. Had they been neophytes it is doubtful they would have been published, and if published they would have been castigated for the unorthodoxy.

 

The Catastrophe versus Uniformity argument had been won by Lyell, not Cuvier a century or so earlier. Woe betide the geologist who stepped out of line and suggested catastrophes might still afflict the globe. Science, eventually corrects this bias, but it takes time for the data to accumulate to a level where it cannot be ignored.

 

Edtharan appears to say this is not the case. That is an error and is thus part of the problem.

[Marat]

Clinical medicine is often practised under constraints that keep it from being fully scientific. Irrational ethical and legal standards derived from the surrounding culture limit the fully rational treatment of the patient; institutional pressures from medical societies further limit rational practice for reasons of promoting the status of the profession, its monopoly over treatment, or its fee schedules. Constraints of time and resources -- since something always has to be done to treat the patient before you now -- also force practice to be less than fully scientific.

 

However, there is also an aspect of medicine commonly called 'medical science,' which is the material found in medical journals, and I think this fits the definition of science. It follows the scientific method; theories are designed to match empirical results and predict new data; falsification of theories is officially cultivated (Popper's criterion); every effort is made to weave the explanation of new facts back into the existing explanatory paradigm, rather than extend it with each new discovery; theory construction operates on a continuum with sciences such as chemistry, physics, biochemistry, organic chemistry, physiology, statistics, etc.

 

But like all sciences, even medical science resists new data which pose a serious challenge to established theories. It does this essentially by denying attention to recalcitrant facts, undermining the reputation of those who discover them, inventing theoretical problems with the experimental methods applied, etc., but all of this is done with much more energy than it would with respect to new data which confirm existing hypotheses. Some of this can be characterized as just a scrupulous, scientific response to something new and startling, but certainly not the tendency simply not to notice, take seriously, publish, or think about data which challenge the existing hypothesis too radically.

 

There was an earlier thread here which exemplified this problem. A scientist at Cornell had conducted an excellent study recently which supported the reality of ESP and there was considerable resistance to publishing it, not for any flaws in its methodology, but just because the result was too startling to be accepted as true.

 

Diabetology is now facing a similar refusal of the field even to notice serious challenges to existing hypotheses, such as the clear evidence now in about 70 scientific journal articles that the complications of diabetes can be prevented -- not by potentially lethal and extremely difficult efforts to normalize blood sugar -- but by taking a simple pill with no reported side-effects, Benfotiamine. Michael Brownlee published important findings showing this in the February, 2003 edition of Nature: Medicine, but most endorcrinologists and diabetologists in North America have still eight years later never even heard of this treatment. The reason is that the Benfotiamine results have never been subjected to large-scale studies, such as the strict blood sugar control hypothesis was during the 1990s in the DCCT study. But why not, given that Benfotiamine could revolutionize and greatly simplify treatment, as well as avoiding 4% to 6% of patient deaths from iatrogenic causes? I think it is just a case of the existing paradigm protecting itself by refusing to mobilize its energy to study any results that refute it, so that it can then always claim to persist unchanged on the argument that the contrary hypothesis "has never been adequately proved." The partisans of Benfotiamine treatment have to agree that the evidence supporting it is insufficient, but not because there is anything questionable about Benfotiamine in itself, but rather, just because the resources will never been made available to back it up. The fact that you see television ads every five minutes for the billion-dollar blood sugar measurement meter industry, which would be nearly wiped out if strict blood sugar control were to be shown to be unimportant by Benfotiamine therapy as an alternative, might well have something to do with it, since no one is going to put up comparable funds to test a drug like Benfotiamine which cannot be patent-protected.

[Edtharan]

I think you have missed the point of my posts entirely.

 

and this sort of proves it:

This is why, for example, Adams' study -- even though if true it would have revolutionary significance -- is still being neglected.

My point is that this has been known for some time and is already being researched. So someone (Adams) publishing a paper on it is not going to be "revolutionary".

 

Yes, he might have shown that the effects are more significant than they were first thought, but that is not a revolution. It also means that the research already under-way will continue as it has (although immunotherpy research might get more funding).

 

Your argument is that science is unwilling to accept radical information. But all you present is an example of non-radical research as proof.

 

I also pointed out that medical research and development takes decades, so even if Adams' research is accepted immediately and was revolutionary, it could be decades before they can actually do anything about it .

[Ophiolite]

Are you contending that their is no inertia to scientific theories that delay a change over to a more productive paradigm? That seems to be the position you are adopting. I would welcome a clarification.

[Marat]

When Ed says that "there is research already being done to correct the autoimmunity of diabetes," this phrase seems a bit vague in its reference. Type 1 diabetes was already thought to be primarily an autoimmune disease in the mid-1960s, and by the mid-1980s efforts were underway with the new neurotonin inhibitor class of immunosuppressives to prevent the development of full-blown type 1 diabetes in children at the very earliest stage of the disease.

 

So this is inded old news, but what is truly new and revolutionary is the theory that the same autoimmune processes that cause the beta cells of the pancreas to be destroyed at the onset of the disease also persist throughout the course of the disease. It had been accepted before that that the autoimmunity burns itself out once the beta cells are destroyed, but Dr. Faustman's work in the 1990s in trying to reverse type 1 diabetes in the NOD mouse model revealed that pancreatic beta cells continue to try to regrow, but are continuously being destroyed by an autoimmunity which never ceases. This first raised the possibility that this regrowth is also occurring in human type 1 diabetes, but is also being stopped by a continuing autoimmunity. If this were so, then interventions to suppress the autoimmune processes, such as are routinely used in other autoimmune diseases like lupus, might also allow the regrowth of beta cells to get ahead of the destructive processes.

 

But with Adams' work and a scattering of similar papers in other areas of diabetology, we are now entering a third phase of our understanding of diabetes and autoimmunity, since the truly revolutionary discovery here is that the continuing autoimmunity may not only be destroying the pancreatic beta cells, but also causing the late sequelae of the disease -- the vascular and neurological deterioration -- which had always been thought, and is now still thought by more than 99% of clinicians and scientists working in this field to be caused by hyperglycemia. If treatment were now to switch from the usually futile, extremely burdensome, and often lethal effort to control hyperglycemia to using extemely toxic immunosuppressive drugs to control the autoimmune processes really causing diabetic complications, then this would truly be a clinical and scientific revolution. It would involve scientists in the field becoming so convinced that autoimmunity was really the cause of complications that it was worth accepting all the severe downsides of immunosuppressives, such as heightened cancer and infection risks, cataract formation, and atherosclerosis.

 

Adams himself explicitly recognizes that this would be a revolutionary development, since in his abstract he states: "If this [autoimmune hypothesis of the etiology of diabetic complications] is so, the therapeutic implications are immense, involving a switch from ineffectual tight glycemic control to immunotherapy."

 

As I said above in my discussion of Benfotiamine, of course there are other reasons to treat diabetic hyperglycemia, so as to avoid hyperosmolar coma, predisposition to infection, vitamin loss through polyuria, ketoacidosis, and to maintain growth and metabolism. But the type of control of hyperglycemia in this case would be radically different from the extremely dangerous and burdensome type of control now practised, and could certainly be designed to eliminate neurologically damaging and often lethal hypoglycemia. I'm not really sure what Edtherian is saying in response to this by commenting that "because of the loss of function of the pancreas the people with diabetes still end up with low blood sugar." Of course he must mean high blood sugar, but if this is treated with insulin not to normalize glucose levels but just to avoid the more extreme conditions reviewed above, low blood sugar could largely be eliminated as the huge clinical problem it is today.

 

I'm also not sure what is meant by Rocket's comment that medicine is not a science. Just because it is a science which is directed to practical outcomes, and in its response to practical problems is not always scientific, this does not mean that it is not a science in its development of its theory structure and its testing of that structure by reference to empirical data. Just because people often drive badly and irrationally, and thus in an 'unscientific' way, does not mean that the principles of mechanics that go into designing cars are not a science. There are even MMS degrees -- 'Master of Medical Science' -- given by many universities, so those arguing that medicine is not a science should take up their objects with those august institutions.

 

With respect to what I am saying about scientific revolutions and the resistance of established explanatory paradigms to recalcitrant evidence, all of this is so widely known, so commonly accepted, and so non-controversial ever since Thomas Kuhn's epic-making 'Theory of Scientific Revolutions' was published in 1974 that I can't believe that people are so unfamiliar with it. There is a whole academic industry built around this field, and if you google its common terms of art, such as 'normal science,' 'abnormal science,' 'scientific revolution,' 'paradigm shift,' etc., you will find countless entries. You can dispute whether 50% of the content of every Ph.D. program in the history and philosophy of science in the world is valid or not, or whether the theory of scientific revolutions makes sense or is empirically well-founded, but my applying it in this context certainly should not be regarded as simply betraying an utter lack of understanding about what science is or how it operates.