Starboy

Arg! Scrambles, don’t put words in my posts. You seem to think that when you apply Ockham’s razor that one is supposed to choose the simplest theory. Yet there are definitions and descriptions of methods that simply say that all Ockham’s razor implies is a bias towards the “simpler” explanation being more likely to be correct. Yet as far as I know, no one has ever done a scientific study of real life scientific choices of two competing theories predicting the same thing but the “simpler” theory being more likely to be the correct one. History has shown that over time a more complicated theory replaces the simpler theory just about every time. In any case scrambles, there are definitions that contradict what appears to be what you think Ockham’s razor is. What makes you think that you understand it and are using it correctly?

Starboy

Scrambles

Starboy,

Simpler = least number of assumptions.

Unnecessary assumptions are dangerous because there are no guarantees that the assumptions are correct. Hence choosing a theory with the least amount of assumptions minimizes the danger of being incorrect.

The reason I think I understand it is because defining complexity the way you seem to define it creates a completely absurd concept. Rejecting theories because they are complicated to explain is silly. The other concept (least amount of assumptions) is not so absurd, and I think you agree with me on that point.

If the intended concept was so absurd we wouldn't know of anything called Occam's Razor. We do know of Occam's Razor, so I'm inclined to think it's not absurd.



Scrambles

Starboy

scrambles, how do you know if an assumption is unnecessary?

Scrambles

Well, I guess that if one theory relies on a subset of assumptions of another theory and both explain the data equally well, then the set of extra assumptions of the second theory were unnecessary assumptions.


Scrambles

Starboy

scrambles, you seem to be implying that Ockham's razor is nothing more than discounting theories that have assumptions that have no effects on the predictions of the theory. If this is so then Ockham's razor is trivial. I don't need Ockham's razor to tell me that one wrinkled dollar plus one new dollars is equal to two dollars is the same as one dollar plus one dollar is equal to two dollars and that introducing the fact about the dollar is wrinkled has nothing to do with how currency is computed. I thought you said that you read the links I posted. This version of the dictum was considered too trivial to be considered important. Scrambles, do you know what you are talking about?

Starboy

Scrambles

Starboy,

Two theories can make the same predictions positing different mechanisms with one theory using a subset of assumptions of another theory.

The extra assumptions can make for a totally different mechanism for explaining observed facts.

Extra assumptions can make profound differences to a theory even though the theory still explains observed data, e.g. Einsteins cosmological constant.


Scrambles

Jayhawker Soule

Noone would suggest that the day of the week has relevance to General Relativity or M-Theory. Excuding the irrelevant is not what parsimony is about.

Perhaps the champions of Occam's Razor could suggest a couple of recent examples of its fruitful application.

Starboy

Scrambles, the more you post the more I think you have no idea what you are talking about. The case of the cosmological constant is an example of the failure of Ockham's razor as you claim to understand it. The less assumptive theory did not turn out to be the better explanation. You are making my case! Keep this up and you will be agreeing that Ockham's razor is of little use.

Starboy

Jayhawker Soule

Interestingly, Popper takes issue with what he terms the "conventionalist concept of simplicity" and notes, instead:

Starboy

ConsequentAtheist, but even in that case it still misses the point. The reasons why one theory should be accepted over another should have nothing to do with simplicity if they to not predict the same things. The deciding factor would then be experiment on nature. Only in the rare case where two theories predicted the exact same things but had different mechanisms or constructs of explanation might one use Ockham's razor. But there again, it would seem to me such a choice would boil down to aesthetics. And in any case if they did posit different mechanisms or constructs it should lead to a program to try to identify those mechanisms or constructs in nature rather than relying on Ockham's razor. I don't see parsimony to be anything more than a subjective aesthetic evaluation.

Starboy