Tercel

Malaclypse,

You do that. And no, I don’t want to get into a flame war with you - I’ve got better uses for my time.
What is your problem here anyway? You said you found Koy’s writing and logic impressive, whereupon I responded as above. I was being honest: I have found Koy to be the most insulting and obnoxious poster on this board. With the result that I, and most of the rest of the Christian posters here no longer respond to him. What is “this shit” about that?

However “chance” is an effective cause – insofar as it can be said that random chance is the thing that is causing the constants to take the values they do or that they take their values by chance.
Hence I see little point in discussing “causes”, as it’s only additional effect seems to be to water-log me in a discussion of whether causality is valid always or whether the universe is caused or uncaused etc.

What on earth is P(E)??? I’m getting tired of seeing this and not knowing what you mean by it.

Okay, I can agree with this.

If by P(E) you in fact are meaning P(E|C) (Since you say above that they’re equal then I assume you are) then I agree. This being the entire design argument in a nutshell: P(E|C) is tiny, and unless you’re so biased as to assume P(D) equally tiny then we <strong>conclude design</strong>.

While this is factually true as it stands (obviously), it is quite irrelevant. Your implication that this shows the FT argument as circular is clearly false though, because in the FT argument are not assuming that “chance is less probable than design” therefore design, but rather: “design has probability X”, scientific information gives chance probability Y, Y is very much less than X, therefore design.

Yes, P(D) is arbitrary. Some are rather more sensible and appropriate than others though.

Each individual must obviously make up their own minds about what they think the a priori probability that the universe was created by a god is. I can say that I think your suggest value is stupid though if I want to.

Yup. Arbitrary doesn’t equal bad though.

Which does not follow at all from the above.
If you’re prepared to accept P(E|C) at 10^-50 and P(C) and P(E|D) as so close to 1 as making no difference (as you seem to do above), then the entire Fine-Tuning argument can be formulated as follows for our readers, given that the number of <a href="http://www2.hawaii.edu/suremath/jsand.html" target="_blank">grains of sand</a> is approx 7.5*10^18:

Do you think the probability that the universe has an intelligent creator is more than the chance of 3 people independently selecting (at random) the same grain of sand given all the beaches in the world to choose from?

If the answer is yes, then the Fine-Tuning argument succeeds.

Given this formulation, I think the merits of the FT argument become readily apparent.
Tercel

[ March 03, 2002: Message edited by: Tercel ]</p>

Malaclypse the Younger

Tercel

This statement speaks for itself. <img src="graemlins/banghead.gif" border="0" alt="[Bang Head]" />

[ February 27, 2002: Message edited by: Malaclypse the Younger ]</p>

tronvillain

Tercel:
Nothing but bias could lead one to assign P(D)&gt;P(E/C) a priori.

Malaclypse the Younger

Tercel

Koy is my friend; I don't like people insulting him; more importantly, your opinions about Koy are irrelevant to this thread and serve only as a distraction.

I don't make that argument in the causal version of the FTA, so this point is irrelevant.

Again irrelevant; I don't make this argument in the causal version of the FTA. You are merely throwing sand in the bulls eyes.

It's helpful, if you want to discuss the FTA, to understand the elementary interpretation of terms in probability theory. P(x) is the a prioiri probability of x, given no additional information. It is usually the size of x relative to the size of the total space in which x appears. When applied to the FTA, P(E) is the size of the life-friendly constant space relative to the total constant space. It is actually exactly the same thing as saying P(E|C), the chance that the universe is life-friendly if we know the constants were generated by chance.

As noted, this is nonsensical. Since the assumption of any value of P(D) is arbitrary, it is nonsensical to infer bias for any given number; or rather all numbers--high or low--can only be assigned by some a priori bias. To say that your numbers are "fair" and mine "biased" is rationally unjustified.

Pardon!? To show that one's conclusion necessarily matches one's assumption is very relevant.

If probability X above had any independent rational justification, you would be correct. However, it does not--it is simply arbitrary and is thus relative only to the the number we know, probability Y.

Unless you can define and rationally justify the appropriateness and sensibility of some numbers over others, this statement is simply false.

If you can prove it's stupid, that would be fair. However, your opinion of the stupidity of a number is irrelevant to a rational argument. Wasn't something said earlier about someone being insulting and obnoxious? Mote, beam, eye.

This is an example of innumeracy. It's impossible to deal with very small numbers like this intuitively. The point is that we have no rational way of determining whether the probability of the existence of a designer is or is not more or less probable than your grain of sand analogy. It might be more probable, it might be less probable. Since the assignment of this probability is arbitrary we have--by definition--no way of distinguishing between arbitrary probabilities like 1/10, 10^-18, 10^-50, 10^-150 (Dembski's number), 10^-150, an enormous Ramsey number like 1/(10^^10^^10); all of these numbers are equally arbitrary.

And calling one number somehow more or less "sensible" or "appropriate" assumes falsely that human beings can deal intuitively with numbers this small.

Essentially you are dressing up an argument from incredulity and innumeracy with some equations. The concept of an event with a probability of 10^-50 happening by chance is simply "incredible", so you search for a designer.

But think, rather, of a deck of cards. Shuffle that deck and examine the contents. The chance of the cards appearing in that order are 8x10^-67. And that's just one deck of cards. The probability that all decks of cards ever played (lets call this a billion) occurring in the order that they did (or any other order) is som unimaginably small number (1/((52*10^8)!))

No one can deal with these tiny probabilities intuitively. Indeed, people cannot deal with even ordinary probabilities intuitively; hence the success of casinos.

Kenny

Actually, this is not quite right.

P(E) = P(C)P(E/C) + P(~C)P(E/~C)

In other words, the initial probability of E has to take into consideration the probabilities that E could occur either by chance or by “not-chance” (however “chance” is being defined).

In fact, to say that P(E) = P(E/C) is to assume that P(E/C) = P(E/~C)

Proof:

P(E/C) = P(C)P(E/C) + [1-P(C)]P(E/~C)

P(E/C)[1-P(C)] = [1-P(C)]P(E/~C)

P(E/C) = P(E/~C)

God Bless,
Kenny

*Edited to Add

Actually, I left out the solution where P(C) = 1, which is another possible way that P(E) could be equal to P(E/C), but I figured you didn’t intend to be that dogmatic.

[ February 28, 2002: Message edited by: Kenny ]</p>

Malaclypse the Younger

Kenny

Well, I guess you're right. The way Tercel has set up the problem, it holds P(E|C) at 10^-50, and P(E) doesn't seem to have any well-defined numerical interpretation; P(E) = P(E|C)*P(C) + P(E|D)*P(D) + P(E|?)*P(?) = ?.

My understanding of the ordinary interpretation of the probability function is that P(x) is defined to be the probability that x will occur by chance. For instance P(royal flush) is the probability that, absent other information, you will receive a royal flush on a fair deal of five cards, computed by dividing the number of royal flushes by the number of total hands possible.

While I suppose that Tercel's formulation is not really "wrong", it requires a nonstandard interpretation of probability theory, which appears at least obfuscatory.

The underlying problem is that Tercel has not set up the problem in a determinable manner. His formulation:

P(E|D) * P(D) &gt; P(E|C) * P(C)

Is not determinable because we have no way of rationally assigning any values to P(E|D), P(D), and P(C). Regardless of the interpretation It is clear, however, that if we set P(E|D) and P(C) arbitrarily close to 1, then the outcome of the inequality is directly dependendent on the assumption of the relationship between P(D) and P(E|C); since there is no rational way to determine P(D), any conclusion (for chance or design) is trivially circular.

I guess Tercel's particular formulation is at least useful for identifying and isolating the circularity.

Clutch

Would someone mind explaining to me how we're supposed to respect the principle of normality in quantifying possible ways the universe might have been, but isn't? I can see having a distant shot at it on the assumption that we hold fixed the actual physical laws. But if even those are up for grabs, if the universe might have obeyed the laws of schmantum schmechanics or fantum fechanics or whatever, and if the masses of the elementary particles might have taken any real number as their values, then how do we quantify the way things might have been in order to produce some principled calculation of the prior probability of things being like this?

Synaesthesia

Clutch,

Or the possibility that numbers and probability theory are meaningless in some "possible" world. There's simply no way of evaluating this kind of thing, I suggest we don't even try.

Wizardry

[Double Post]

[ February 28, 2002: Message edited by: Wizardry ]</p>

Wizardry

P(E|C)

How exactly do we arrive at a value for P(E|C)? How do we reach the conclusion that P(E|C) can have any value at all?

P(E|C) seems to be predicated on the idea that there is a range of values that the physical constants might have assumed. But, is that even possible? Is there any scientific evidence that the physical constants could hold different values than their current ones? If so, is there any scientific evidence of what the possible range of those constants might be? Without firm scientific evidence establishing the both the range of possible values for the physical constants and the range of values required for life, any value of P(E|C) is inherently arbitrary and speculative and therefore worthless.

If there is no valid evidence that the physical values could possibly hold different values then the Fine-Tuning argument fails entirely. P(E|C) becomes 1 and the discussion is over.

Peace out.