Jump to content

Has Ockham's Razor become blunt in the last 700 years ?


studiot
 Share

Recommended Posts

Quote

So allegedly wrote William of Ockham around 1320 AD

Entia non sunt multiplicanda praeter necessitatem

Translated this reads

Things should not be multiplied beyond what is required.

 

Yet no one has found this exact text in his writing that survives.

It is true however that much of his writing echoes this sentiment.

 

Howsoever it maybe the question is as in the title

How true actually is it ?

 

I posted this because members often appeal to this maxim for support.

Link to comment
Share on other sites

Fundamentally I think it does not matter who said it, it is just an attractive way to select among alternative hypotheses without having more data. The reductionist approach allows us to test least amount of parameters first and add more if it does not conform to the experimental outcome.

In principle it is also related to the concept of falsifiability in science. We generally are not able to falsify all potential alternatives, thus we try to start with a minimal framework that we can test rather than going for more complex ones, if they do not have more explanatory power.

However, I am not sure whether it is really a matter of true or not, there is no formal logic (that I am aware of) that underlies it. Rather it is one of the useful rule of thumbs that you can apply in complex situations. 

Link to comment
Share on other sites

HI everyone.  Obviously I quite like Ockhom's razor and a bit of reductionism but I'll play the part of the opposition for the moment.  Thanks for the history, studiot, that was interesting.

Has Ockham's Razor become blunt in the last 700 years ?    Maybe it's not been made blunt enough.   In theoretical physics there is a drive to develop unified theories and also to identify the most fundamental structures and assumptions required to create what we observe around us.  Such a theory has value and may also be fine for a computer to understand and utilise - but it's unliekely to provide understanding to a typical human being. 

   There is value in science that focuses on providing utility rather than prioritising making the least assumptions.  Ockhom's razor made sense back in a time when science was frequently mixed with superstition but in this modern age it seems less relevant.  Since it is still used as a judgement criteria for scientific theories it is having a detrimental effect on the motivation of new science.

    Ockhom's razor still has a place but, as indicated by CharonY, the demonstration and verification of predictions should now be a more important requirement in the judgement of a piece of science (and the inevitable motivation for science that will come from that).

Link to comment
Share on other sites

It is a good guide and often will be correct - but the evidence is the ultimate guide and complex explanations can be the correct ones. "It's the sun" is a lot simpler for explaining global warming than atmospheric back radiation and reduced IR from upper troposphere to space, but will be incorrect.

Link to comment
Share on other sites

Any model selection process which uses AIC or BIC (or any other variable penalisation method) is explicitly using Ockham's razor to prevent over-fitting of models. I'd wager their use is increasing not decreasing.

Link to comment
Share on other sites

12 hours ago, Col Not Colin said:

HI everyone.  Obviously I quite like Ockhom's razor and a bit of reductionism but I'll play the part of the opposition for the moment.  Thanks for the history, studiot, that was interesting.

Has Ockham's Razor become blunt in the last 700 years ?    Maybe it's not been made blunt enough.   In theoretical physics there is a drive to develop unified theories and also to identify the most fundamental structures and assumptions required to create what we observe around us.  Such a theory has value and may also be fine for a computer to understand and utilise - but it's unliekely to provide understanding to a typical human being. 

I'm not sure why that matters in general, or to the applicability of Ockham's razor.Whether they typical human being understands some bit of science is not why we pursue the science.

 

12 hours ago, Col Not Colin said:

   There is value in science that focuses on providing utility rather than prioritising making the least assumptions.  Ockhom's razor made sense back in a time when science was frequently mixed with superstition but in this modern age it seems less relevant.  Since it is still used as a judgement criteria for scientific theories it is having a detrimental effect on the motivation of new science.

I think the concept applies to more than science, and is not anywhere near a full description of the motivation of new science. Yes, you might do science to find a simpler explanation, but people also do new science because current science is incomplete, or possibly wrong, and to realize experiments of theory where the technology has caught up, or to push the limit of what experiment can do. Those have nothing to do with having the simplest explanation. 

There are places where it has been applied. We had some data suggesting superluminal neutrinos some years back, but the simplest explanation was experimental bias/error, and that's precisely what it turned out to be. The more complex explanation - that there was new physics - was not going to be adopted without a whole lot of confirmation, because of another adage: extraordinary claims require extraordinary evidence. Which, if you think about it, is just another version of Ockham's razor (much like we have multiple ways of stating the second law of thermodynamics)

 

Link to comment
Share on other sites

Ockham's-razor rule of thumb rests on two simultaneous optimisation desiderata: 1) Maximum simplicity. 2) Fitness to account for observation.

The search for maximum simplicity works under the constriction to fit experimental data. The latter overrides it all. If explanations seem more complicated it's likely because the range of phenomena that we intend to contemplate is widening more than ever before. Further constrictions operate on approximations, ancillary hypothesis, etc., to account for an ever more complicated landscape of phenomenology.

I tend to agree with the points as expressed by @CharonY, @Ken Fabian, and @Prometheus even though I cannot be totally sure that we would completely agree with each other in the finer details.

Summarising, I think Okham's razor is alive and well, even though it's become subtler and more difficult to apply it.

Edited by joigus
Link to comment
Share on other sites

46 minutes ago, joigus said:

Summarising, I think Okham's razor is alive and well, even though it's become subtler and more difficult to apply it.

I think the OP has a point though, perhaps not when applied to science but philosophically speaking; the more complicated our world becomes, the more we resort to a simple explanation i.e. the elite's are trying to make it complicated. And that blunts our/most people's understanding. I think the philosophy behind the intention/knife edge applies to anyone who wants to actually understand this world/universe; and how many of us actually want's that, rather than the simplest explaination as to why I shouldn't even try?

 

Link to comment
Share on other sites

1 hour ago, joigus said:

Summarising, I think Okham's razor is alive and well, even though it's become subtler and more difficult to apply it.

Like Pullman's Subtle Knife perhaps ?

4 hours ago, Prometheus said:

Any model selection process which uses AIC or BIC (or any other variable penalisation method) is explicitly using Ockham's razor to prevent over-fitting of models. I'd wager their use is increasing not decreasing.

Thank you so much for introducing a modern area of statistics I had never heard of   -  off to investigate!     +1

 

15 hours ago, Col Not Colin said:

Thanks for the history, studiot, that was interesting.

Since you are interested his actual words were

Quote

Frustra fit per plura quod potes fieri per pauciora

Which translates as

It is futile to employ many principles when it is possible to employ fewer.

But he did not say that the extra principles are untrue

and all too often it is used as an excuse to reject the (more) complicated, whether justified or not.

We are all a bit lazy and like to use the excuse "neglecting (higher order) terms as insignificant"

This applies to my favourite engineering formula "something squared over twice something else"  , which is not validated by the razor but by mathematics.

But I would love to see the razor applied to the theory of flight of eg a human made mechanical aircraft,  a butterfly,  a hummingbird and  a hot air balloon.

Link to comment
Share on other sites

3 minutes ago, studiot said:

Carry on ?

I thought I had...

38 minutes ago, dimreepr said:

I think the OP has a point though, perhaps not when applied to science but philosophically speaking; the more complicated our world becomes, the more we resort to a simple explanation i.e. the elite's are trying to make it complicated. And that blunts our/most people's understanding. I think the philosophy behind the intention/knife edge applies to anyone who wants to actually understand this world/universe; and how many of us actually want's that, rather than the simplest explaination as to why I shouldn't even try?

 

 

Link to comment
Share on other sites

2 minutes ago, dimreepr said:

I thought I had...

Are you saying that the razor suggests Archimedes principle  - which is of course an incorrect explanation ?

Edited by studiot
Link to comment
Share on other sites

Just now, studiot said:

Are you saying that the razor suggests Archemedes principle  - which is of course an incorrect explanation ?

No, I'm replying philosophically; I understand Archemedes principle - I don't understand why it's incorrect...  

Link to comment
Share on other sites

2 minutes ago, dimreepr said:

No, I'm replying philosophically; I understand Archemedes principle - I don't understand why it's incorrect...  

If the air pressure inside the balloon was not marginally greater than the air pressure outside the balloon would collapse inwards.
 

Since the pressure is greater, the density must be greater and thus the mass of displaced air is less than the mass of the air inside the balloon, even without the basket and trappings.

This is different from a rigid balloon such as a hydrogen or helium one where there is a positive bouyancy force.

Link to comment
Share on other sites

1 minute ago, studiot said:

If the air pressure inside the balloon was not marginally greater than the air pressure outside the balloon would collapse inwards.
 

Since the pressure is greater, the density must be greater and thus the mass of displaced air is less than the mass of the air inside the balloon, even without the basket and trappings.

This is different from a rigid balloon such as a hydrogen or helium one where there is a positive bouyancy force.

You seem to be arguing against your premise in the OP, since you placed the question in the "general philosophy" forum.

Scientifically, the razor is almost always valid, because it can be reduced to a binary question; that's seldom true in life...

Link to comment
Share on other sites

5 minutes ago, dimreepr said:

You seem to be arguing against your premise in the OP, since you placed the question in the "general philosophy" forum.

Not at all. You picked out a clear cut case where the 'obvious' simple explanation is incorrect.
The actual reason is more complicated.

In the OP I am posing a discussion question, has the razor been superceded by modern science ?

My answer is obviously therefore yes, since there is at least one case where it is unreliable.

But it is a good discussion and it is my hope that when (new) members produce outlandish 'theories' and say it must be true because of Ockham's Razor this thread can be pointed to as evidence of that unreliability.

Link to comment
Share on other sites

On a minor note, how are we all identifying "the least number of assumptions"?

A Theory in Biology can use a small number of words and what appears to be a small number of assumptions.  However, the moment any of that is something like "and this keys to the recptor site", there are actually a thousand-and-one small assumptions attached.  There are assumptions that microscopic objects like molecules behave like macroscopic objects.  That ping-pong ball models of molecules actually convey information about shape.  Within each molecule the inter-atomic bonding is also represented by some generalised theory of bonding.  In particular, it was never assumed necessary to find exact solutions to any Schrodinger's equation for the molecule.

If Ockham's razor is applied as a judgement tool for science then Chemistry isn't far removed from wishful thinking and Biology is something akin to superstition.   However, Biology is obviously an extremely valuable science.  That is the one that has developed vaccines for the Covid-19 virus.  It has demonstrability of predictions and utility for human beings "in spades" and that is what makes it a high value science.

If we had been determined to make the least number of assumptions, then computers would still be trying to find numerical solutions to wave equations today and human beings would not yet have had the understanding that molecules in the Pfizzer  or  AstraZeneca vaccine could even hold together.

Link to comment
Share on other sites

3 minutes ago, Col Not Colin said:

On a minor note, how are we all identifying "the least number of assumptions"?

Good question, how many giant's shoulders are our assumptions standing on ?  +1

Link to comment
Share on other sites

1 hour ago, Col Not Colin said:

If Ockham's razor is applied as a judgement tool for science then Chemistry isn't far removed from wishful thinking and Biology is something akin to superstition. 

I think the implication is that it is based on the assumptions within a given theoretical framework. I.e. we are comparing biological hypotheses with each other for the most part. And even there, a molecular question will have different competing hypotheses than organismal or ecosystem-wide ones.

Link to comment
Share on other sites

Just to add another perspective, neural networks are being used to model increasingly complex phenomena and in general the more parameters the model has, the better they perform - so long as it has sufficient data on which to learn. I mean, most of them have more hyper-parameters than normal models do parameters.

Link to comment
Share on other sites

8 hours ago, studiot said:

Like Pullman's Subtle Knife perhaps ?

Sorry, I'm not familiar with it.

9 hours ago, dimreepr said:

I think the OP has a point though, perhaps not when applied to science but philosophically speaking; the more complicated our world becomes, the more we resort to a simple explanation i.e. the elite's are trying to make it complicated. And that blunts our/most people's understanding. I think the philosophy behind the intention/knife edge applies to anyone who wants to actually understand this world/universe; and how many of us actually want's that, rather than the simplest explaination as to why I shouldn't even try?

I agree that the OP has a point, even when applied to science. But I still think it all has to do with the scope of what you want to explain, with what we could call first principles vs particular explanatory pathways based on those principles. In the spirit of what @Prometheus says, there are overarching principles (simple), and then there is the implementation of particular scenarios (complicated parametrics). Something like that.

I want to make more comments. Perhaps later. I need some sleep. The discussion is tantalizing.

I feel a bit behind the game right now. Maybe I'm just tired.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.