What is this scientific principle called?

[HJ Farnsworth]

Greetings,

This is more of a question about the philosophy of science than physics. In fact, it's just a vocabulary question, so should have a one-word or one-phrase answer (similar to "Occam's razor"), if there is one.

Is there a word or phrase for the following scientific principle (my phrasing):

"If multiple theories describe observed data equally well, then they must all be regarded as plausible/possible underlying explanations for the data, (regardless to how intuitive they seem relative to each other)."

When I say, "equally well", above, I really mean, "equally well within the limits of accuracy of the tools used to observe the data." For instance, SR and Newtonian kinematics describe observed data equally well in low-velocity regimes - unless you have sufficiently-accurate measurement equipment, in which case SR describes even the low-velocity regime stuff better. If you don't have such accurate measurement equipment, though, the principle described above basically says that you have to regard both theories as possibilities, even though SR is a much less intuitive explanation for most people.

Thanks for any help you can give.

-HJ Farnsworth

 

[turbo]

Is this a homework question? I could give you a one-word suggestion, but I am reluctant to do so in this context.

 

[HJ Farnsworth]

Hello,

Thanks for the response. This isn't for homework - though I don't know how I could prove that, so I understand if you don't want to answer.

On occasion, a conversation with friends or just stuff I'm thinking about results in me wanting some piece of vocabulary that I don't have, and I sometimes go to physicsforums to find the vocabulary because I don't really know what to look up to find it (I had a previous thread that was similar to this one where the answer ended up being "qualia", for instance).

That's what this post came from - I want to know the term simply because it seems like it would be a nice concept to know the term for.

 

[turbo]

OK, look up parsimony, to see whether a simpler explanation might be preferred over a more complex one. That might steer you into some good areas.

 

[DiracPool]

I think the word you may be looking for is "equipotent."

 

[Borek]

Funny, wikipedia redirects parsimony to Occam's razor.

 

[DiracPool]

Funny, wikipedia redirects parsimony to Occam's razor.

Well, they are related but Occams razor refers specifically to theory whereas parsimony is more of a general-purpose concept, so they are not interchangable terms.

 

[256bits]

Would 'rival theories' work for you.

 

[jedishrfu]

How about M-Theory for when 5 theories are four too many? :-)

 

[Studiot]

Diracpool
I think the word you may be looking for is "equipotent."

Nah, the word is equipollent.

 

[HJ Farnsworth]

Thanks for the responses.

I've been digging around the suggestions provided here. Here's what I found:

Parsimony is related to Occam's razor, and the two terms are sometimes used interchangeably, though they are different (I think the latter may be a specific instance of the former). In fact, one of the definitions I found of parsimony was, "economy in the use of means to an end; especially : economy of explanation in conformity with Occam's razor". The idea with Occam's razor is that given two theories, the one that makes the fewest assumptions should be regarded as more likely.

Equipollence and equipotence are synonyms - the former term is more widely used. One definition I found of the term is purely mathematical - two sets are equipollent if there is a bijection between them. A definition more relevant to this discussion that I found is, "Equipollence - equality between two or more propositions, as when two propositions have the same meaning but are expressed differently".

These are both related to what I was looking for, but not quite the same. The principle that I'm looking for actually competes with parsimony/Occam's razor, in that it advocates inclusiveness of multiple theories as possible explanations, whereas parsimony gives criteria to eliminate certain theories.

Equipollence (the second definition above) has to do with equivalent theories, expressed differently/with different postulates. An excellent example of this is the numerous postulates which could be substituted for Euclid's parallel postulate in Euclidean geometry, resulting in identical theories. This is a special case of the principle that I am looking for - it does not include the case of theories that are not consistent with each other, although they could still predict similar results. For instance, SR and Newtonian kinematics are fundamentally non-equivalent theories, though they predict convergent results at low velocities.

I think the term that I am looking for is underdetermination, http://en.wikipedia.org/wiki/Underdetermination. Wikipedia's definition: "In the philosophy of science, underdetermination refers to situations where the evidence available is insufficient to identify which belief we should hold about that evidence." Later in the same Wikipedia article, "To show that a conclusion is underdetermined, one must show that there is a rival conclusion that is equally well supported by the standards of evidence."

 

[Studiot]

I'm sure the digging is worth doing and sharing.

After all fancy words are of little use unless both sides of a discussion understand them and agree on their meaning.

So well done in your effort to promote wider understanding.

 

[HJ Farnsworth]

Thank you very much!

 

[DiracPool]

Thank you very much!

Hold on Farnsworth, not so fast. Before you start congratulating yourself, it's a good idea to ask yourself (and tell us) why you are looking for this term. That didn't seem to flesh out in the discussion so far, and it could make a difference. Is it a class assignment, or you're using it to write an assigned paper for school? In that case underdetermination might be your word, since instructors many times don't like students who get too fancy. However, if you are writing for yourself or a scholarly journal/publication, you may want to make a statement and invent your own word. This is perfectly acceptable depending on what your writing about, I do it all the time in the neuroscience field I publish in.

It's your party so I don't have any suggestions for you, all I can say is that if you do create a word be sure to state that that is what you are doing, define it as clearly as you can (as you have done here), and put the word in quotes the first time you use it. In other words, don't just start using it without stating what you are doing. I personally might do this in your case because "underdeterminism" sounds a bit flat, and if it isn't precisely what you mean, create a new term. It also looks cool, your readers will respect you for your initiative.

 

[HJ Farnsworth]

Hold on Farnsworth, not so fast. Before you start congratulating yourself, it's a good idea to ask yourself (and tell us) why you are looking for this term.

Heh heh not to worry, consider the champagne bottle unpopped. I'm not really sure why I started looking for this term. It's not really for anything official, like homework or publications - more just because it popped into my head that it would be a useful term to know for talking about science with people.

It's hard to pinpoint, but the reason that the search popped into my head was probably some combination of...

1. Studying science history, which I have been doing lately for amusement and which is intricately related to the philosophy of science.
2. Tutoring cousins and friends in various math and science subjects. It's often useful and a nice change of pace to talk about science as a whole after having worked on something more rigorous.
3. Random pseudo-philosophical conversations with friends (surely, we've all done this one?).

Doing these things, I guess I thought, "it would be nice to have a word for blah blah blah [see my first post]".

So, I may be cliche/corny for saying intellectual curiosity - but I'm going to say intellectual curiosity.

 

[DiracPool]

I'm going to say intellectual curiosity.

Nothing wrong with that. I'd say that after dabbling in several fields, what I am really is simply a general "philosopher of science." I'm an origins guy, I want to know why things are the way they are fundamentally, classically even, simply and elegantly. So I can explain them to and discuss them with non-scientists, family, children etc. That's what really interests me. So I know where you're coming from. I just fell into a neuroscience specialty cause I feel that is where such a quest needs to begin, mind is principal, primary.