Is Occam's Razor Just a Cliche in Today's Complex World?

[DanP]

Is Occam's razor becoming nothing more than a poor cliche those days ? To be more exact, I am referring here in special to the it's more naive version for which internet seems very fond, quoting it in many different flavors time and again ... "The simplest explanation is the most plausible one".

After all nature couldn't care less about the fact we like simple things. It proved to be complex and weird. Quantum mechanics put an end to IMO to the belief that nature is simple.
String theories introduced us to worlds with 10, 11 , whatever dimensions.

The correct theory from a pool of theories with equal qualities may prove time and again not to be the simplest one. Yet the cliche still stands.

(of course, I am not referring here to the fact that Occam's razor has 0 value as a scientific result. but only the philosophical aspect)

 

[DaveC426913]

Is Occam's razor becoming nothing more than a poor cliche those days ? To be more exact, I am referring here in special to the it's more naive version for which internet seems very fond, quoting it in many different flavors time and again ... "The simplest explanation is the most plausible one".

After all nature couldn't care less about the fact we like simple things. It proved to be complex and weird. Quantum mechanics put an end to IMO to the belief that nature is simple.
String theories introduced us to worlds with 10, 11 , whatever dimensions.

The correct theory from a pool of theories with equal qualities may prove time and again not to be the simplest one. Yet the cliche still stands.

(of course, I am not referring here to the fact that Occam's razor has 0 value as a scientific result. but only the philosophical aspect)

I don't think you're really understanding Occam's Razor. You seem to think it simply means "nature is simple". And when nature turns out to not be simple, you think Occam's Razor is non-applicable.

 

[rewebster]

Is Occam's razor becoming nothing more than a poor cliche those days ? To be more exact, I am referring here in special to the it's more naive version for which internet seems very fond, quoting it in many different flavors time and again ... "The simplest explanation is the most plausible one".

After all nature couldn't care less about the fact we like simple things. It proved to be complex and weird. Quantum mechanics put an end to IMO to the belief that nature is simple.
String theories introduced us to worlds with 10, 11 , whatever dimensions. The correct theory from a pool of theories with equal qualities may prove time and again not to be the simplest one. Yet the cliche still stands.

(of course, I am not referring here to the fact that Occam's razor has 0 value as a scientific result. but only the philosophical aspect)

they are still theories though

 

[leroyjenkens]

I think that it keeps being used so many times, it gets simplified more and more until the real purpose of it is lost.
It's like that game where one person whispers something to another person, then that person whispers it to the next, and so on until they get to the last person and by the time it gets there, the message is sometimes completely different from what it was originally.

 

[bp_psy]

"The simplest explanation is the most plausible one".

After all nature couldn't care less about the fact we like simple things. It proved to be complex and weird. Quantum mechanics put an end to IMO to the belief that nature is simple.

Do you know of any working explanation of physical phenomena at the atomic level that is more simple than quantum mechanics?

 

[DanP]

Do you know of any working explanation of physical phenomena at the atomic level that is more simple than quantum mechanics?

No. Did Einstein knew one when he stated "God does not play dice" ? |I don't know, but this never prevented him from never being really at ease with the probabilistic nature of quantum mechanics. He basically rejected Occam's Razor in this case (or embraced it, probably thinking that something simpler than a probabilistic nature must be at work) and continue to pursue something he never found, developing further his own theories.

IMO the moment in time most applicable for this rule of thumb was back then , in the first quarter of the last century. From a long time we don't really need heuristics and rules of thumb for QM, we believe its true because the outstanding predictions it made and the phenomenal accuracy with which they where verified.

Anyway I found a nice paper which discusses all I wanted to know. It has a mini-chapter on the myth of simplicity and simplicity criteria which are interesting to read.

http://www.scientificexploration.org/journal/jse_21_1_gernert.pdf

 

[bp_psy]

No. Did Einstein knew one when he stated "God does not play dice" ? I don't think so, but this never prevented him from never being really at ease with the probabilistic nature of quantum mechanics. He basically rejected Occam's Razor in this case (or embraced it, probably thinking that something simpler than a probabilistic nature must be at work) and continue to pursue something he never found.

Occam's razor does not put any limitations on the existence of a simpler explanation. It is also does not make any assertion concerning the nature of the physical world.So if the phenomena explained by quantum mechanics could me explained by a simpler theory with the same results then by occam's razor the simpler one probably the real explanation.The problem is that two theories that would always give the same result would be indistinguishable so saying that one is the real theory is nonsense anyway. The usefulness of Occam's razor comes out in a situation like the following: If prediction of a theory is contradicted by 100 experiments then the theory is probably wrong. If 99 good experiments give the same result and 1 contradicts them then that 1 experiment was not done properly.

 

[DanP]

The problem is that two theories that would always give the same result would be indistinguishable so saying that one is the real theory is nonsense anyway.

Sure, its probably the case with Lagrangian formalism and Newtonian formalism. Good point.

The usefulness of Occam's razor comes out in a situation like the following: If prediction of a theory is contradicted by 100 experiments then the theory is probably wrong. If 99 good experiments give the same result and 1 contradicts them then that 1 experiment was not done properly.

What do you think about a case like in which exist more theories, each perfectly anchored in the science of the day (i.e not crackpot stuff) , both sane mathematically and consistent, with a different degree of complexity, but none of those able to do any predictions whatsoever which can be verified by experiment. (like the situation that we don't have predictions which can be verified today from String Thories).

Does it worth invoking Occam's razor and focus on the less complex one ?

 

[madness]

Is Occam's razor becoming nothing more than a poor cliche those days ? To be more exact, I am referring here in special to the it's more naive version for which internet seems very fond, quoting it in many different flavors time and again ... "The simplest explanation is the most plausible one".

After all nature couldn't care less about the fact we like simple things. It proved to be complex and weird. Quantum mechanics put an end to IMO to the belief that nature is simple.
String theories introduced us to worlds with 10, 11 , whatever dimensions.

The correct theory from a pool of theories with equal qualities may prove time and again not to be the simplest one. Yet the cliche still stands.

(of course, I am not referring here to the fact that Occam's razor has 0 value as a scientific result. but only the philosophical aspect)

Occam's razor advises us to choose the simplest expanation which is consistent with our observations. If the simplest explanation is very complicated, that's ok as long as there isn't a simpler option. And of course if we can find further information to distinguish between explanations then we have no need to use Occam's razor. You're right that from a pool of theories the correct theory may not be the simplest, but we are advised to choose the simplest one if we have no other means to distinguish them.

However, I would say that other factors such as the aesthetic appeal of a theory probably come into play in real life. And the simplicity of a theory seems like a subjective measure which people may disagree on.

 

[Char. Limit]

Don't cut yourselves on Occam's Razor.

Sometimes the simplest explanation really is the best. For example, the simplest explanation is that both DanP and Dave are annoyed at each other, but both had good intent.

 

[DaveC426913]

Is Occam's razor becoming nothing more than a poor cliche those days ? To be more exact, I am referring here in special to the it's more naive version for which internet seems very fond, quoting it in many different flavors time and again ... "The simplest explanation is the most plausible one".

There will always be people who misuse platitudes. Why dwell on it?

The correct theory from a pool of theories with equal qualities may prove time and again not to be the simplest one.

Has that happened? If not, why are you borrowing trouble?

 

[DanP]

There will always be people who misuse platitudes. Why dwell on it?
Has that happened? If not, why are you borrowing trouble?

Actually, Wikipedia states on the Occam's razor entry that there where cases in which the rule delayed progress. An example was given : "...It originally rejected DNA as the carrier of genetic information in favor of proteins, since proteins provided the simpler explanation. "

I do not know in how many instances this happened or how founded the claim is. I'll read this night the paper I linked in my previous posts, since it appears (from my cursory glance) to contain a larger analysis of this.

But I do not think that this is very important (how many times happened, if at all) from a strictly philosophical point of view.

As the current science theories become more and more complex and:

- it will be often very difficult to decide which is the "simplest" one, because perceiving a thing as simple is highly subjective. two experts at the top of their field may very well see 2 different theories as "simple and beautiful". It wouldn't be unheard.

- today several theories are mere mathematical models, and can't benefit yet from the advantage of making predictions which can be confirmed experimentally

In those conditions Occam's razor is still useful or becoming a cliche. I was curious what ppl think about it.

It's actually not borrowing trouble. It's discussion for the sake of the discussion, hence it was posted in a philosophy section.

 

[Skyhunter]

Occam's Razor, or the principle that "plurality should not be posited without necessity", is an often abused concept on the internet.

But it is not that the simplest hypothesis is the correct one, but that the hypothesis that makes the fewest assumption is more likely to be correct. This is usually the simplest, but not exclusively.

 

[JTE]

Calling something a cliche is not the same as saying it is untrue. Worth separating these two. Occam's Razor is definitely cliche, so call it Parsimony and give it some fresh legs for a while. Politicians are good at this sort of trick.

As for whether it's true, it's certainly the simplest explanation of why it has so often been observed to be true.

 

[Dale]

Occam's razor advises us to choose the simplest expanation which is consistent with our observations. If the simplest explanation is very complicated, that's ok as long as there isn't a simpler option.

This is also my understanding of Occham's razor.

 

[elect_eng]

The problem with Occam's razor is that it has not kept up with the times. In the latest TV commercials I see that razors now have 5 blades. Simple things become complicated over time. Perhaps a future version of Occcam's razor will say the physics laws tend to fit into one of the first 5 simplest explanations.

Seriously though, what does this principle really do for us? It's human nature to keep things as simple as possible. It's human nature to seek out a simple pattern buried in a confusing situation. Sometimes we even think we see a simple pattern when it is not really there. The principle is basically useless. We do science with the scientific method. Anything important automatically falls out of that process.

 

[arildno]

Actually, Wikipedia states on the Occam's razor entry that there where cases in which the rule delayed progress. An example was given : "...It originally rejected DNA as the carrier of genetic information in favor of proteins, since proteins provided the simpler explanation. "

Why was this "delayed progress"??

At that time, all you could find out about DNA wa that it was, largely,...INERT.

The idea that the nucleic acid had, for example, som structural, stabilizing role was at least as consistent with available data than that it was the carrier of genetic information.

So, I must ask you, in what way did that opinion "delay" progress?

Was it lack of progress to decode many proteonic interactions?

Did the idea prevent development of technology that could be used to determine the true role of DNA?

 

[Gokul43201]

As for whether it's true, it's certainly the simplest explanation of why it has so often been observed to be true.

 

[russ_watters]

Why was this "delayed progress"??

...Was it lack of progress to decode many proteonic interactions?

I'll take it a step further. If we're looking for a method which overall provides the most benefit for the least effort, then it doesn't matter if there are individual examples where it led us down the wrong path.

 

[arildno]

I'll take it a step further. If we're looking for a method which overall provides the most benefit for the least effort, then it doesn't matter if there are individual examples where it led us down the wrong path.

Agreed.

A theory that needs to get 10 assumptions proven right is more research expensive than one needing only to prove 5 assumptions, UNLESS those 10 happen to be vastly easier to verify...

 

[noblegas]

Whats the big deal? So Occam's Razor does not apply in all situations where you try to find the simplest explanation to a particular problem , big deal. It is only a rule of thumb just like Newton's law of gravitation(which probably should be called a law since it does not apply in all physical situations). It is not like the scientific method where the scientific method is your only method at your disposable that is used as a means of discovering new phenomena about the physical world.Occam's razor is a principle based on the application of logic, and even though it was created like in the forthteenth century , the greeks like Aristotle still applied it , at least in its prototypical form, and obviously it failed then because it turned out that Aristotle's claim that their are only 4 elements in the universe and that heavier objects fall faster were dead wrong, even though they were simplistic explanations.

 

[DanP]

I'll take it a step further. If we're looking for a method which overall provides the most benefit for the least effort, then it doesn't matter if there are individual examples where it led us down the wrong path.

No doubt, this is a good argument.

 

[mgb_phys]

The problem with Occam's razor is that it has not kept up with the times. In the latest TV commercials I see that razors now have 5 blades.

Maybe Occam's razor would be a good name for a single bladed razor?

Of course Occam's razor is also the reason for believing that airmail is delivered by giant homing pigeons.
1, For any given pigeon you see the odds that this is the world's largest individual are vanishingly small.
2, Therefore there will always be a larger pigeon
3, Therefore you can have arbitrarily large pigeons to carry Fedex packats.

The alternative is to imagine the millions of people that would be required to build some sort of mechanical flying machine and the places it would take off and land. Manage the whole operation, inspect the operation, manage the inspecting of the managing of the operation - and oversee the managing of the inspection of the managing of the operation. It's completely ridiculous.

 

[Tao-Fu]

Yes, I still find Occam's razor to be a good principle. It simply urges you not to multiply entities in your theory beyond need: Don't insert superfluous agents into the model. If you can explain the same phenomena with less machinery then do that for your own convenience! If another theory explains more phenomena or explains the same phenomena better then Occam's razor doesn't even come into the discussion. I would invoke it any time two different theories perform a satisfactory job. I would opt for the simpler (in my view) description.

As I see it, it goes along with model agnosticism. We can never know which model is true, so use the model that you find most convenient for the current problem.

 

[elect_eng]

Yes, I still find Occam's razor to be a good principle.

It is a good principle to use in foresight, but it's so obvious that it seems unnecessary to even talk about it.

In hindsight, Occam's razor looks incredibly stupid. If we look back at theories for any phenomenon, we see more and more complex theories replacing simpler ones. This trend will continue for sure, yet we keep applying Occam's razor rather that trying to make intuitive leaps ahead of a need illuminated by new experimental evidence.

Maybe it would be better to create a new principle.

Occam's guillotine: "The simplest theory is surely wrong, so devise new experiments that will provide evidence that shows the limits of your current theory, and give you a reason and direction to find a more complex one."

Well, I guess that one is obvious too. Like I said before, the scientific method takes care of all we can do as limited humans. These types of principles are nice to package in a catchy phrase and quote to the layman, but otherwise they are redundant elaborations on the scientific method and common sense.

 

[DaveC426913]

No.

I've modified your statement:

...we see more and more complex theories replacing simpler ones as new entities are discovered and thus necessarily added.

Occam's Razor discourages duplication of unnecessary entities.

 

[elect_eng]

No.

I've modified your statement:

Occam's Razor discourages duplication of unnecessary entities.

Of course that is a perfectly acceptable (even though implied) change. If they were duplications of unnecessary entities, they would not even be part of our scientific history. But, it was not the discouragement from Occam's razor that guided these progressive changes. It was the scientific method that was the agent that demanded these changes as necessary steps. And, the look backward reveals the uselessness of Occam's razor as an agent for forward progress.

 

[DaveC426913]

Of course that is a perfectly acceptable (even though implied) change. If they were duplications of unnecessary entities, they would not even be part of our scientific history.

You seem to be thinking that these unecessary entities get propogated throughout our collective body of knowledge. I am thinking they only occur when first forming a theory. Of course they don't end up in history. Nobody records theories that are rejected before they get wide acceptance.

And, the look backward reveals the uselessness of Occam's razor as an agent for forward progress.

Occam's Razor allows us to make a choice when two theories of otherwise equal validity are on the table - before they are accepted. It is not a tool for progress; it is a tool for breaking a stalemate. A stalemate does not last long enough to make the history books.

 

[Char. Limit]

I always thought that Occam's Razor was not "the simplest theory is likely to be the correct theory", but that it was "the simplest theory consistent with all observations is likely the correct theory".

The reason we dumped Newtonian physics was because it wasn't consistent with observations at very large or very small scales.

 

[DaveC426913]

I always thought that Occam's Razor was not "the simplest theory is likely to be the correct theory", but that it was "the simplest theory consistent with all observations is likely the correct theory".

Yes. Geez, I hope no one was saying otherwise...

I think all the paraphrasing we're doing here is leaving a lot out but is nonetheless definitely implied.