Atheists against Occam's Razor

[Taverner]

I am an Atheist. And yet, when it comes to Occam's Razor I do not believe.

I do not believe in taking the most likely answer. In fact, as a believer in chaos theory, I find the razor to be incomplete.

Sometimes things DO happen out of whack. The smart observer will take note of anomalies, but in the end has to have a clean slate in his or her mind before rendering judgment.

[Orrex]

1. i hadn't realized that Occam's Razor had anything to do with 'the most likely answer'

Most parsimonious, sure. But most likely? That's a new spin on it for me.


Must consider...

[Donald Ian Rankin]

2. ...Chaos theory?

"Chaos theory" is a term that gets overused nearly as much as "quantum" or "second law of thermodynamics".

I'm currently doing a PhD in chaos theory. It's a branch of pure mechanics, dealing with dynamical systems with certain rather technical properties (generally, a set I with a transformation T: I -> I and two subsets J1, J2 c I such that T(J1) = T(J2) = I, although there are other definitions).

It's not something that tells you anything about theology.

Also, if there's a theorem that proves that reanimated dinosaurs will inevitably escape, it got left out of the courses that I studied

I'm sorry to be so hostile - I can see where you're coming from - but you're making one of my my bete noire mistakes, and Michael Crichton isn't around for me to rant at.

[Taverner]

3. But the end result is the true random sequence

It proves chaos theory

The true random sequence exists in math - we're just not willing to admit it

[Donald Ian Rankin]

4. What are you talking about?

"It proves chaos theory"?

Chaos theory is an area of research, not a single result. It contains results that have been proved, and unsolved questions, but there is no single "chaos theory" to be proved or disproved.

Maths contains all sorts of notions of true random sequences. No-one denies that. Deterministic chaotic systems, however, do not generate them.

[Taverner]

5. Checking out at this point

Not. Being. Useful.

[rug]

6. I didn't expect this.

[Random_Australian]

9. Huh, we here in physics have something we call "chaos theory", but we can afford to be

sufficiently fuzzy about it that we can ignore that it is not a theory at all, and if we want to explain it we can ignore most of the mathematics as well. :) Of course, any reasearch into the field has to be more rigorous, but when talking about it to people I usually just use

"There are systems in which, given two identical starting states, any difference will exponentiate"

And leave it at that. :)

[Random_Australian]

10. Oops, had that wrong, but editing period had expired.

x( It just struck me that I'm not referring to chaos theory proper at all, but something that would be better termed "a property that physical systems can have referred to as chaos or chaotic behaviour"

[trotsky]

7. I'm not sure you understand what Occam's Razor really is.

It doesn't mean "take the most likely answer."

Classically it's stated "entities must not be multiplied beyond necessity." In other words, if something doesn't add any explanatory value, it should be discarded. Gods fall under this description.

[Random_Australian]

8. The razor is incomplete? What the hell are you talking about?

Ok, let's run through an example.

I take some measurements of the length of a rod versus temperature.

I get a series of five dots that looks a lot like a straight line. Of course, there are always error in measurement so it is never an exact fit.

I fit a straight line, and find R-square = 0.998 (A very good fit)

Then I fit a quintic polynomial, because it has enogh degrees of freedom that I can always fit it to five points exactly.

Lo and behold, the match is exact, unlike the straight line.

Do I have evidence that it is not a straight line?

According to Mr. Occam, not really. I always could fit the curve to five points exactly. No matter what they were I could always fit them exactly, so I know nothing about them from the fact that I fitted a quintic polynomial to it.

And this is the heart of Occam's Razor - don't add meaningless things.

And you say that not only do you not believe it, but that the idea that the differences between systems can exponentiate contradicts this.

I wonder what your reasoning is.

[skepticscott]

11. Occam's Razor is often stated badly or misunderstood

but it isn't really a matter of believing it or not. It's simply a rule of thumb for rational inquiry, not a law of nature.

To say that "the simplest explanation is the best" really doesn't cover things. The "simplest" explanation at any particular point may not, in the final evaluation be the correct one. I prefer to state it as "When trying to explain something, start with the simplest explanation that covers all of the facts and work from there."

But in fact, science really IS about finding the best and most likely explanations for things, and that approach is fully capable of accounting for anomalous observations. If you have a mountain of evidence that supports a theory and a few observations that seem to contradict it (as is not unusual when doing real-world science), it's reasonable to ask which is more likely...that the theory is true and that the few contradictory observations are mistaken or unreliable, or that the theory is false and that a far greater amount of evidence is all somehow flawed.

[Occam Bandage]

12. In order for me to begin to formulate a counter-argument,

I would have to know what the hell kind of twisted proto-thoughts were colliding in your brain when you posted this.

[cosmik debris]

13. I also see Occam's Razor differently.

When I learned about it several decades ago it was stated like this: "The most probable answer is the one with the fewest leaps of faith."

Therefore, the probability of Yahweh, Zeus, or FSM actually existing is identical because they require the same number of leaps of faith.

You may "take" whichever one you want, but you do not have probability to support your argument.