Thursday, February 19, 2009

On Randomness

I suspect this might get deeper in philosophy than I'm used to, but I'll go ahead and lay out my ideas: I think "random" is pretty much just an admission of ignorance. When I roll a die, the result isn't "random" because there's some metaphysical principle coming up with utterly unpredictable numbers. It's "random" because I didn't bother to collect detailed data on the die and my throwing technique necessary to predict the outcome beforehand. It's "random" because of a lack of data.

If I were to hand the die to Lt. Commander Data, he could probably weigh the die, feel the variation in density, and perform enough fine motor control to roll whatever he wanted. In fact, he did so in one episode at a craps table. The result is no longer random because he has all the information he needs to predict the roll under different circumstances and use his fine motor control to pick out the result he wants by matching the circumstances.

That's pretty well why I've been getting vocal about calling Creationism for what it is: Randomness. Whenever we find something that wouldn't make sense for an intelligent designer to do, they start making claims that we can't possibly know anything about the stone idol behind it all, and thus can't make predictions about its behavior.

Sounds just like the randomness I described.

12 comments:

King of Ferrets said...

Yeah, if we have enough data we can probably predict anything "random".

Except maybe QM. That's a slippery little thing. Though it's also likely that we just don't know the variables that control what happens.

Lifewish said...

I'd say there's two definitions of random. The first is "conformant to a statistical distribution rather than a deterministic rule". QM falls nicely into that category: photons are predictable en masse, just not individually. Many pseudorandom phenomena like dice are sufficiently unpredictable that they're best treated as random in this sense.

The second definition of random is "lacking any ground rules or limits". A die that was random in this sense would be as likely to spontaneously achieve escape velocity or turn into a small chicken as it would be to show a 6.

The first kind of randomness is scientifically useful, because it allows concrete predictions to be made (albeit to a certain confidence level rather than in absolute terms). If your hypothesis predicts something at the 99.999% level and it doesn't pan out, that's a hefty strike against your hypothesis.

The second kind is not scientifically useful. An hypothesis is an exclusion of certain possible outcomes. Since an all-powerful entity can achieve any outcome, it is by definition impossible to produce testable hypotheses about it. The situation is even worse when the entity is intelligent, because then it could be actively trying to deceive you (e.g. omphalos creationism).

I'm currently studying statistics for actuarial exams, so it's kinda important to me not to confuse the two (or I might find myself writing "goddidit" in place of answers on my exam sheet...)

Bronze Dog said...

Kind of dips into an earlier post: The Creationist type of randomness is less predictive than regular randomness.

Anonymous said...

From what little I understand of quantum mechanics, there are elements that don't just appear random but are intrinsically random by their very nature, that is, if it WAS in fact predictable, that is, potentially knowable in advance if one knew the right variables, the results would be different than what certain experiments observe.

That doesn't apply to a dice roll or my computer's random seeding algorythms though, and it certainly has nothing to do with evolution.

MrFreeThinker said...

“[E]volution works without either plan or purpose … Evolution is random and undirected.”
(Biology, by Kenneth R. Miller & Joseph S. Levine, pg. 658 (4th edition, Prentice Hall, 1998); emphasis in original)

Bronze Dog said...

Right. One probably out-of-context quote somehow negates the fact that natural selection and many other forces in evolution are non-random, and bypasses this discussion of what random means.

Stop thinking like a keyword search bot.

Bronze Dog said...

To put it another way, evolution is stochastic. You have some "random" variables, being mutations that are then acted on by orderly forces. The results won't be exactly the same each time you run it, but the results are often quite predictable.

Of course, people like MrFreeThinker would love to use the equivocation to think this is a bad thing, even though such processes have useful results. I'm going keep using Kriging, no matter how much he tries to call it random.

Anonymous said...

MrFreeThinker:

What emphasis was in the original? See, when you cut and paste that from Luskin or wherever else you found it, you forgot to add some.

Dear oh dear.

Furthermore, what is omitted by the ellipse? Do you know?

The official purpose of this website is ... to ... hit people ... if they have the odd desire to search for my name.

I am ... confused ...


Casey Luskin, caseyluskin.com/facts.htm.

See how easy it is to make people sound like they said something they didn't if you don't have the context of the original text. You do have the original text, don't you? You can show us that nothing important was missed out by the ellipse, can't you?

By the way, when will we be seeing your proof that Terence is metaphysically impossible? The burden of proof is, by your own admission, with you after all.

Tom Foss said...

Also, I'm sure a single quote from a 10-year-old introductory textbook contains all the detail, accuracy, and nuance that exist in the theory.

Please, go on to tell us what the rest of the chapter says, MrFreeThinker. And then maybe take a 200-level textbook out, and see what it says.

Don said...

I seem to remember some quote-mine from a book on evolution (may have even been Origin of Species) that surfaced a few years ago where an ellipsis stood in for most of the book.

If I remember correctly, it took part of a sentence from the first chapter, inserted the "..." and then part of a sentence from a chapter at or near the end of the book.

That's one hell of an ellipsis.

Valhar2000 said...

I think I remember that one. It was a 9 chapter elipsys, the longest ever made by a creationist, according to PZ Myers.

I don't think it was from The Origin, tough.

Lifewish said...

“[E]volution works without either plan or purpose … Evolution is random and undirected.”
(Biology, by Kenneth R. Miller & Joseph S. Levine, pg. 658 (4th edition, Prentice Hall, 1998); emphasis in original)


So are you putting that forward as your understanding of evolution, or as some lame attempt to claim that that's what everyone believes about evolution, or what?

This may be surprising for you, but skeptics do not invest a lot of faith in authority figures. Rather, we try to understand the arguments that these authority figures use to support their beliefs, and try to judge whether these arguments are valid.

So presenting us with an arbitrary (mis)quote from an arbitrary "evolutionist" will not have the same effect as (e.g.) presenting a Biblical literalist with a Bible quote that contradicts them. The quote will only be significant if you are willing and able to defend its conclusion against informed dissent.

Are you?