Flaw in the Fine Tuning Argument

[Dogbyte]

The way you commonly use Bayes' Theorm, doesnt appear to have very much meaning, since you just plug in any claim as if they are immune to normal examination. If the claim doesnt pan out, then obviously so does your probability function. What is illogical about that?

Im disputing that you are offering up any chance hypothesis, looks more like one of necessity, but i'll leave it up to you defend your own assertions.

[cnearing]

I'm using likelihood ratios.

As for what's illogical about that, it's gibberish.  You're not ayibg anything that ven means anything to anyone who knows probability theory.

If you prefer to call my hypothesis a necessity hypothesis, that's fine.  It still defeats the FTA.  Perhaps more embarassingly so.

However, it isn't a necessity hypothesis, any more than the fact that, on theism, the designer is necessary makes theism a "necessity" hypothesis.

[Dogbyte]

It looks like your just complaining that WLC hasnt given *you* enough arguments against your favorite chance hypothesis, even your made up M' theory. Why should anyone be concerned with nothing other than sound examination of the theories on the table based on the background info and evidence? If you think the hypothesis can stand up to scrutiny, then go ahead and show where it lacks.  If you want to argue against P2, then go argue for a hypothesis from necessity or chance that can show to be more reasonable. merely positing M' as you did, is not even a hypothesis, its a bare assertion with nothing offered for support, which begs the question. Is M' plausible based on the evidence? Let us here it.

[Dogbyte]

I'm using likelihood ratios.

As for what's illogical about that, it's gibberish.  You're not ayibg anything that ven means anything to anyone who knows probability theory.

If you prefer to call my hypothesis a necessity hypothesis, that's fine.  It still defeats the FTA.  Perhaps more embarassingly so.

However, it isn't a necessity hypothesis, any more than the fact that, on theism, the designer is necessary makes theism a "necessity" hypothesis.

i'd prefer to call your hypothesis whatever it is you say it is, but you've not really given any reasons for anyone to think anything about it, other than its your made up invention. so if you dont know, i for sure dont either.

[cnearing]

You misunderstand.

I don't  think M is true.  It just defeats the justification for p2, whichbis that the likelihood of fine-tuning on non-design, non-necessity hypotheses is very low.

It isn't supposed to be a serious hypothesis that you should believe (theism isn't a serious hypothesis that you should believe, either, as demonstrated by the fact that it is way worse than M) but rather to show that the reasoning behind p2 is a failure.

And it does that handily.

How about this: can you calculate a reasonable P(|FT|chance) value?

If you do it without considering the essentially infinite hypotheses resembling M, you have failed.

Just like Craig failed.

[Dogbyte]

You misunderstand.

I don't  think M is true.  It just defeats the justification for p2, whichbis that the likelihood of fine-tuning on non-design, non-necessity hypotheses is very low.

It isn't supposed to be a serious hypothesis that you should believe (theism isn't a serious hypothesis that you should believe, either, as demonstrated by the fact that it is way worse than M) but rather to show that the reasoning behind p2 is a failure.

And it does that handily.

How about this: can you calculate a reasonable P(|FT|chance) value?

If you do it without considering the essentially infinite hypotheses resembling M, you have failed.

Just like Craig failed.

Well how else would you expect someone to rationally hold a belief, until a chance or necessity hypothesis is shown to be strong? Justification in that sense is a human property, so you have to convince the human.

What good will it do if anyone shows that some probability calc can ~1 given an infinite number hypotheses (whatever that means,  other than asserting M')? What about the actual state of affairs in *this* universe based on evidence? Its a poor misunderstanding of probabilities to just think you can fling something called "infinite hypotheses" at a string of letters that stand for whatever you choose,  without considering any background evidence for it being actual and weighty. You could argue for an infinite number of things being probable, which would just allow everything on the table, no justification required...which was what you were after in the first place for P2.

If you have some unrestricted infinite hypothesis, that has no need for scientific claims or improbability, that sort of takes the need for justification away from any actual state of affairs doesnt it?  If you have some restrictive hypothesis where a mechanism is required to actualize it, then multiply that by infinity, you just push the problem one step back further, to which i'd ask that it still looks like fine tuning given this special mechanism, and places you back to apparent fine tuning. Not sure what you are advocating other than probabilistic predictions do exist and are useful for some cases, some cases it is not, or in some cases it can be misused, im sure you know of the informal fallacies associated. Why are you shirking any reasoning for your M' or any other chance hypothesis? What they claim matters.

If you want justification, then you'll need to provide some reasons other than multiplying your ideas by infinity. Thats all it appears you use probabilities for, you seem to leave out the stuff we need for justification. 

[cnearing]

What I am saying is simple: Craig attempts to reject a large space of possible hypotheses on the basis that the probability of FT conditioned on that hypothesis space is very low.

The problem is that he never calculates this probability in any meaningful way.

Do you see that much, at least?

[Dogbyte]

What I am saying is simple: Craig attempts to reject a large space of possible hypotheses on the basis that the probability of FT conditioned on that hypothesis space is very low.

The problem is that he never calculates this probability in any meaningful way.

Do you see that much, at least?

I have no idea what you might consider to be "meaningful". Maybe you can elaborate.

Talking about how narrow the life permitting ranges are with respect to different constants and quantities is a useful discussion I think. It doesn't seem to be enough to posit that some universe had to have these narrow conditions, the issue is rather why is a life permitting universe so much more improbable, than a life prohibiting one?

[cnearing]

Before you can ask why something is improbable, you must first establish that it is improbable.

Can you calculate the likelihood on which Craig rests his argument?  Can you find anyone who has?

What, precisely, do you think is "improbable" and why do you think it is improbable?

[Dogbyte]

Before you can ask why something is improbable, you must first establish that it is improbable.

I can agree to that,  but there you will then go on to discuss as to what one means by "establish"  something as improbable.  You are back to looking for reasons for justified belief, so take the claims about all the several constants and quantities being arbitrarily set and independently so from each other,  and ask whats the best explanation based on what we know about these areas.

[cnearing]

Great.  So we have this model, and it has some set of constants and parameters, and these constants and parameters have to have very precise values to make the model match reality. 

If we accept some very strong form of scientific realism, this is the same as saying that the universe itself probably actually constants and parameters of these values. 

So, we ask, either:

A.) Why is the universe such that this model requires such precise values of its constants and parameters to match it, or

B.) Why does the universe have these values for its constants and parameters (related to each other by the form of this particular model)?

The proponent of the fine tuning argument, depending on whether he has taken interpretation A or interpretation B, above, says,

A.) Well, our universe allows for life as we know it, and most universes matched by a model of the same form with different values for its parameters would not. 

B.) Our universe allows for life as we know it, while most universes with different values for its constants and parameters (related to each other, again, by the form of this particular model) would not.

The apologist's task, then, is to take one of these two interpretations and show that he can use this fact to exclude or reject all cosmological hypotheses that do not include the universe being designed, or to show that some set of design hypotheses has a higher likelihood for the fact in question (whichever fact you pick, depending on your interpretation) than any competing set of non-design hypotheses. 

Unfortunately, none of the apologists who actually pursue this line of reasoning possess even the philosophical sophistication to frame the question in a cogent manner, much less the ability to answer it or perform the task that their argument requires. 

Think about the two most prominent FTAs in common usage: Craig's and Collins's.

Craig's relies on the premise "Fine tuning is not due to necessity or chance."  However, his justification for this premise relies on the assertion that "the probability of fine tuning on chance is unimaginably low."  This, however, is either a lie or a sophomoric error.  In fact, Craig has never evaluated this probability at all, and neither has anyone he has quoted.  In fact, he has made it clear that he is too deeply ignorant of probability to evaluate this probability even if he wanted to.  He doesn't actually even understand what the phrase he has used means. 

Collins's argument relies on the premise that "P(LPU|T) > P(LPU|ASU)."  However, while Collins is more or less using likelihoods correctly, here, he also has not evaluated these likelihoods and, indeed, clearly does not possess the command of probability theory to actually do so, or, in fact, even a full understanding of what these terms represent. 


This is the same critical failure in both cases: the argument rests on a likelihood (the probability of whatever "fine tuning fact" you want to consider given some space of non-design hypotheses, be it "chance" or "ASU") but in both cases that likelihood is not calculated.  It is not estimated.  In attempting to discuss it, both "philocophers" reveal that, in fact, they have no idea what it is or how it might be calculated at all.

Until some calculation of these conditional probabilities is actually offered, any rational person who actually understands the matter under discussion must reject both arguments.  They are both laughable failures, serving only to highlight their author's remarkable lack of philosophical sophistication. 

[Dogbyte]

Great.  So we have this model, and it has some set of constants and parameters, and these constants and parameters have to have very precise values to make the model match reality. 

If we accept some very strong form of scientific realism, this is the same as saying that the universe itself probably actually constants and parameters of these values. 

So, we ask, either:

A.) Why is the universe such that this model requires such precise values of its constants and parameters to match it, or

B.) Why does the universe have these values for its constants and parameters (related to each other by the form of this particular model)?

The proponent of the fine tuning argument, depending on whether he has taken interpretation A or interpretation B, above, says,

A.) Well, our universe allows for life as we know it, and most universes matched by a model of the same form with different values for its parameters would not. 

B.) Our universe allows for life as we know it, while most universes with different values for its constants and parameters (related to each other, again, by the form of this particular model) would not.

The apologist's task, then, is to take one of these two interpretations and show that he can use this fact to exclude or reject all cosmological hypotheses that do not include the universe being designed, or to show that some set of design hypotheses has a higher likelihood for the fact in question (whichever fact you pick, depending on your interpretation) than any competing set of non-design hypotheses. 

Unfortunately, none of the apologists who actually pursue this line of reasoning possess even the philosophical sophistication to frame the question in a cogent manner, much less the ability to answer it or perform the task that their argument requires. 

Think about the two most prominent FTAs in common usage: Craig's and Collins's.

Craig's relies on the premise "Fine tuning is not due to necessity or chance."  However, his justification for this premise relies on the assertion that "the probability of fine tuning on chance is unimaginably low."  This, however, is either a lie or a sophomoric error.  In fact, Craig has never evaluated this probability at all, and neither has anyone he has quoted.  In fact, he has made it clear that he is too deeply ignorant of probability to evaluate this probability even if he wanted to.  He doesn't actually even understand what the phrase he has used means. 

Collins's argument relies on the premise that "P(LPU|T) > P(LPU|ASU)."  However, while Collins is more or less using likelihoods correctly, here, he also has not evaluated these likelihoods and, indeed, clearly does not possess the command of probability theory to actually do so, or, in fact, even a full understanding of what these terms represent. 


This is the same critical failure in both cases: the argument rests on a likelihood (the probability of whatever "fine tuning fact" you want to consider given some space of non-design hypotheses, be it "chance" or "ASU") but in both cases that likelihood is not calculated.  It is not estimated.  In attempting to discuss it, both "philocophers" reveal that, in fact, they have no idea what it is or how it might be calculated at all.

Until some calculation of these conditional probabilities is actually offered, any rational person who actually understands the matter under discussion must reject both arguments.  They are both laughable failures, serving only to highlight their author's remarkable lack of philosophical sophistication.

Are you after a statistical type of calculation?

[cnearing]

Some approximation of a probability calculation is required, yes, if the argument is to be accepted.

Specifically, there is a set of "chance" hypotheses, call it C := {c1, c2, ...}

The probability of FT given chance, or P(FT|C), is equivalent to:

P(FT|c1)*P(c1|C)+P(FT|c2)*P(c2|C)+...

Some non-arbitrarily approximation of this term is required.

Craig clearly has no idea how to even approach the problem, and the same is obvious true for Collins.  This is sufficient grounds for any rational person to dismiss their arguments as failures.

If you can do better, let me know.

[Dogbyte]

Some approximation of a probability calculation is required, yes, if the argument is to be accepted.

Specifically, there is a set of "chance" hypotheses, call it C := {c1, c2, ...}

The probability of FT given chance, or P(FT|C), is equivalent to:

P(FT|c1)*P(c1|C)+P(FT|c2)*P(c2|C)+...

Some non-arbitrarily approximation of this term is required.

Craig clearly has no idea how to even approach the problem, and the same is obvious true for Collins.  This is sufficient grounds for any rational person to dismiss their arguments as failures.

If you can do better, let me know.

Have you read Collins defense of the teleological argument?

[cnearing]

Yes, I have.  He doesn't even try to approximate this calculation.

What's worse is that he actually has to do a similar calculation for theistic hypotheses, but he doesn't even try to do that, either.  His approach is conceptually better than Craig's --you should do both calculations--but he fails to do either, just as Craig fails to do the one.