Science and Arguements From Authority

[MaxwellsDemon]

I recently explained to a young teenager that one of the things that makes science great is that you don't have to "take people's word for stuff". Science differs in that if you don't just have to blindly believe and accept dogmatic arguements. You can always just go out and perform an experiement to see the results for yourself. This objectivity and repeatability has always been a major strength for science and sets it apart from authoritative philsophies like those underlying religion. That discussion got me thinking though about whether this is still true in modern science. Nowdays scientific disiplines are such broad fields that you usually have only a few specialized experts with detailed knowledge who are doing research in a particular area. Others almost have to just "take their word for it" because to check their results would require years of training. Also, as we probe deeper into natures secrets, we require increasingly sophisticated and complex machinery. I can't, for example, build my own large hadron collider to double check the results of physicists working there. How do I know that the results aren't just a peculiarity of that particular machine or location? Anyway, it seems to me that modern science is moving toward a philosophy where arguements from authority hold more sway than they used to. What are your thoughts on this matter?

 

[Dale]

I disagree. The scientific method remains the same.

 

[ZapperZ]

I recently explained to a young teenager that one of the things that makes science great is that you don't have to "take people's word for stuff". Science differs in that if you don't just have to blindly believe and accept dogmatic arguements. You can always just go out and perform an experiement to see the results for yourself. This objectivity and repeatability has always been a major strength for science and sets it apart from authoritative philsophies like those underlying religion. That discussion got me thinking though about whether this is still true in modern science. Nowdays scientific disiplines are such broad fields that you usually have only a few specialized experts with detailed knowledge who are doing research in a particular area. Others almost have to just "take their word for it" because to check their results would require years of training. Also, as we probe deeper into natures secrets, we require increasingly sophisticated and complex machinery. I can't, for example, build my own large hadron collider to double check the results of physicists working there. How do I know that the results aren't just a peculiarity of that particular machine or location? Anyway, it seems to me that modern science is moving toward a philosophy where arguements from authority hold more sway than they used to. What are your thoughts on this matter?

The one thing that you need to always keep in mind is that, in this day and age, it is extremely difficult to get a consensus among scientists in a particular field. In other word, the scientists are their own skeptics whenever something new is presented. If someone reports a new discovery, you can bet that someone else will try and verify the validity of that discovery.

This has never changed, and it is even more in practice nowadays, even when the experiment is as big and complicated as the LHC. We can already look at the Tevatron for an example. CDF and D0 are two large detectors at the Tevatron. Often, they're both looking for the same thing, but using slightly different methodology. And there is a very healthy competition going on between the two of them. While one can't say that there is a clear animosity between them, there is that competition to outdo the other, or even try to falsify the result from the other. So when they both agree on something, one often take notice.

What this means is that, when there IS a consensus, you can be assured that that discovery, or that idea, has gone through the wringer many times. Now, this is never a guarantee that that idea is perfectly valid (there is no such guarantee in physics over all possible range), but it means that even if you and I can't actually go out and test it out, someone else (or maybe even more than one) had. One only needs to look at the ongoing tests on various aspects of QM and SR and GR, even though, especially for QM, it is such a well-established theory.

Zz.

 

[humanino]

As technology improves, it becomes cheaper and cheaper to perform checks of older experiments. The first cathode ray vacuum tubes, which are really the first accelerators, could only be built in by a handful of experts in the world. A few decades later, one could easily afford such a particle accelerator at home, in every television set. Nowadays, television sets do not even use cathode ray tubes anymore, and genuine particle accelerators are available commercially. There are research ideas for novel acceleration technologies, and it is not impossible that in the future, tabletop accelerators will have better performances than the most powerful accelerators in the world. There still remain many problems, such as how to detect, acquire, and analyze the decay product of the collisions. Yet one could argue that in a sufficient distant future, LHC-like experiment could be performed as high-school labs !

You may take me seriously or not. Another example is Michelson–Morley experiment types. At the time, those were top world-class research experiments. Nowadays I can buy a laser pointer in my local supermarket, and better experiments are actually performed in high-schools. So my first argument is that our general principles in physics are tested routinely when we educate from kids to future researchers.

My second argument is that it is not true professional researchers do not test general principles. A good deal of high-energy physics consists in searching for physics beyond the standard model of particle physics. There are more accelerator labs in the world than LHC searching for physics beyond the standard model, and there are more strategies than accelerator based physics to go beyond the standard model.

My third argument is that there are sometimes people studying other stuff, applying the standard model of particle physics "as such" (taking it for granted), for their own purposes (not testing this standard model), and they could stumble upon a disagreement. For instance, people studying the Sun will want to detect neutrinos. When we first found an experimental disagreement with our theoretical expectations on the number of neutrinos from the Sun, first and for quite a while we suspected our models for the Sun. As it turns out, the solution was not a misunderstanding of the inner workings of the Sun, it was a deeper misconception in the initial formulation of the standard model that neutrinos are massless.

And I could go on...

 

[wofsy]

My observation of biologists is that false results get exposed sooner or later - usually sooner - because even though research is elaborately specialized there are a zillion specialists in each area.

I think that your point is valid for a different situation, when scientific research is politicised. When this happens the scientific community has been known to accept certain premises with inadequate scrutiny and has tended to believe what is wants to rather than demanding rigor.

Examples in my mind are:

- The Eugenics movement in which scientists bought off on the premise that allowing "inferior" people such as "retards" or Jews to reproduce was creating a genetic crisis and would if not destroy the species at least create unbearable suffering in future generations. Scientists and academic institutions flocked to study this problem. They did this proactively on their own initiative and also sought the huge funding that was made available for research.

- The anti-nuclear energy movement in Western nations smacks of the same fear based science.

When this happens it does become difficult to check results. Data is often not generally available. Critics are vilified. Funding is not available for research that disagrees with the premise.

This sort of thing reminds me of the use of science to support dogma in the Middle ages where the Ptolemaic system was declared absolutely true and critics e.g. Gallileo were vilified.

Interestingly, both Eugenics and the Ptolemaic system contained elements of truth but they were used to squelch open inquiry and to achieve social goals. I am sure the fossil fuel comapnies just love all of that anti-nuclear energy stuff coming out of the scientific community - though here again there is truth in it.

 

[madness]

When doing my degree in physics I feel like I'm continually having to accept things from authority. For example, in quantum field theory, I just have to accept that if I have an infinite ground state energy, it's ok to just subtract it. I also accept from authority that something cannot be true and false at the same time - just like people from different cultures have accepted that something can be true and false at the same time.

 

[humanino]

When doing my degree in physics I feel like I'm continually having to accept things from authority. For example, in quantum field theory, I just have to accept that if I have an infinite ground state energy, it's ok to just subtract it. I also accept from authority that something cannot be true and false at the same time - just like people from different cultures have accepted that something can be true and false at the same time.

You are saying that you have to accept authority arguments concerning purely theoretical statements. I'm sorry but this is bad logic : those are basic axioms of the physical theory. There are a certain number of axioms to quantum field theory, including elementary logic (where things are not true and false at the same time). What you have to accept but also have to opportunity to challenge whenever you please is the claim that these axioms describe experimental results.

So, you are welcome to come up with a different theory, based on different axioms, and prove that it is compatible with what we know and measure. Other people try to do it all the time.

You will also note that the internal mathematical consistency of the standard model of particle physics, in terms of pertubative renormalizability of quantum field theories, is much better understood and established today than what one can read in most textbooks. I can elaborate if you are interested.

 

[madness]

Being allowed to subtract infinity from an equation is an axiom of quantum field theory? And my point was that elementary logic is learned from authority, which is clear from the fact that other cultures did not assume the law of excluded middle. I don't doubt that these theories agree with experimental results, its just that they come to these results through logic which I would not find acceptable if I weren't told they are acceptable.

 

[humanino]

Being allowed to subtract infinity from an equation is an axiom of quantum field theory?

No it is not axiom, but it is a calculation rule which can be deducted from the axioms and proven to be mathematically consistent.

And my point was that elementary logic is learned from authority, which is clear from the fact that other cultures did not assume the law of excluded middle. I don't doubt that these theories agree with experimental results, its just that they come to these results through logic which I would not find acceptable if I weren't told they are acceptable.

t'Hooft for instance has proposed a microscopic model of quantum mechanics based on non-deterministic cellular networks. It is a basic construction from the theory of cellular automaton that non-deterministic networks are equivalent to deterministic ones, which makes this construction quite interesting. So, yes I believe there are people trying to construct models based on non-standard logic instead of ranting about having to accept current models.

 

[madness]

Can you provide any sources which explain why it's acceptable to subtract infinity from an equation in QFT? And I understand that it is possible to "question authority" in science, it's just that the process of learning it (at university at least) involves continually accepting things from authority.

 

[humanino]

Can you provide any sources which explain why it's acceptable to subtract infinity from an equation in QFT?

Sure.
Feynman integrals and motives

This article gives an overview of recent results on the relation between quantum field theory and motives, with an emphasis on two different approaches: a "bottom-up" approach based on the algebraic geometry of varieties associated to Feynman graphs, and a "top-down" approach based on the comparison of the properties of associated categorical structures. This survey is mostly based on joint work of the author with Paolo Aluffi, along the lines of the first approach, and on previous work of the author with Alain Connes on the second approach.

Algebra for quantum fields

We give an account of the current state of the approch to quantum field theory via Hopf algebras and Hochschild cohomology. We emphasize the versatility and mathematical foundation of this algebraic structure, and collect algebraic structures here in one place which are either scattered over the literature, or only implicit in previous writings. In particular we point out mathematical structures which can be helpful to farther develop our mathematical understanding of quantum fields.

Both are proceedings, not quite "lectures" but basically in the style of reviews.

And I understand that it is possible to "question authority" in science, it's just that the process of learning it (at university at least) involves continually accepting things from authority.

Everybody has their own learning methods. When it comes to theoretical physics, Poincare and Feynman both advocated to first read a result, then second try to prove it yourself, and then only third compare your proof to whoever you got the result from. I believe both did it consistently from University to the end of their career, or claimed to do so. If today you think this is an impossible strategy (which in itself is debatable), I believe this is still something one should at least give a shot at as often as possible.

 

[madness]

"Everybody has their own learning methods. When it comes to theoretical physics, Poincare and Feynman both advocated to first read a result, then second try to prove it yourself, and then only third compare your proof to whoever you got the result from. I believe both did it consistently from University to the end of their career, or claimed to do so. If today you think this is an impossible strategy (which in itself is debatable), I believe this is still something one should at least give a shot at as often as possible."

That's what I do in maths courses, but the physics lectures don't really work that way since they aren't based on theorem/proof. The kinds of problems I feel I have to accept from authority are e.g why it's acceptable to move poles off the axis in contour integrals just because it gives a physical answer. I'm sure some of the problems stem from a lack of knowledge.

 

[JoeDawg]

Anyway, it seems to me that modern science is moving toward a philosophy where arguements from authority hold more sway than they used to. What are your thoughts on this matter?

Its not just science, modern society is all about specialization. Mostly because there is just too much information for anyone person to know, even within a field of study. So like any other field, science suffers from this problem. However, science does offer the ability, that given someone who does have the technical skill and knowledge, results can/should be repeatable. Not entirely satisfying...


This might interest you:

http://plato.stanford.edu/entries/scientific-knowledge-social/#BigSciTruAut