Ok Stak,
I’ve read the Sober article, well the beginning of it.
Sober says: This paper defends two theses about probabilistic reasoning. First, although modus ponens has a probabilistic analog, modus tollens does not – the fact that a hypothesis says that an observation is very improbable does not entail that the hypothesis is improbable. (I won’t quote the second theses since you only used the first.)
What he is saying is that if you have a deductive argument in modus ponens form you can set up a probabilistic modus ponus analog and the probability conclusion will be reliably true. However the same can not be said of a probabilistic analog for modus tollens form..that is if you convert an M.T argument into a probabilistic M.T …the probabilistic conclusion can not be relied upon.
So let’s look at some of the things you wrote;
-No it's not a good argument, it's invalid. That definition of the God hypothesis, it's negation does not follow from Dawkin's six point argument.
- When it comes to logical validity, all we are concerned with is form, not that the premises are true or false. The way the argument is presented, it's not possible for the conclusion (negation of the GH) to necessarily follow from his 6 premises.
-When I say an argument if invalid, I mean that the argument is built in such a way that it contains formal and informal errors that call into question the ingrity of the argument. I have no problem with the content of what he said, just that he did not arrange the argument in such a way that makes it strong.
And
Dawkin's own argument is a simple modus ponens.
A --> -G, A, .:. -G
If the ARGUMENT of this chapter is accepted, the factual premise of religion- the GOD HYPOTHESIS- is untenable (that means NEGATION).
So in other words Marg, if A is sufficient for -G and we accept A, -G will follow necessarily.
Using your same letters,
This is the argument Stak and it’s modus ponens a validity constructed argument:
1) If the premises are true (A) then ….the H God is false (G]
2) The premises are true [A]
Therefore …the H God is false [G]
converting into probabilistic M.P analog
1) If the premises are true (A) then… the H God is highly probably false
2) The premises are true [A]
Therefore the H God is highly probably false.
Note this consistent with what Dawkins argued which was.. If we accept his premises as true then the H God is false. And Sober does say that this is deductively valid in the probabilistic form. In other words if we knew that as long as the premises were true the H. god would be false, then finding out later that the premises are indeed factually true would mean the H. God would be false.
Now if we put this argument into Modus Tolens valid form ..it doesn’t remain valid in probabilistic M.T. (that is what Sober’s is saying)
1) If premises are true [A] then… the H. God is False [G]
2) The H. god is not false[~G]..
Therefore the premises are false. [~A]
That is deductively valid.
However the probabilistic version is not valid.
1) if the premises are true [A] then… the H God is probably false [G]
2) the H god is not false [~G]
Therefore the premises are probably false. [~A]
So even if we knew for a fact that the H. God is not false or in other words that the H. God is true..that does not allows us to conclude reliably that the premises are probably false.
So Stak, Dawkin's argument is deductively valid as far as form goes and even when put into probablistic form it is still deductively valid.
The problem that you have with the argument is that you don't accept that those premises lead to or warrant his conclusion. You think the reasoning of the argument is poor. That it is set up in valid form is a not the issue that concerns you, it is the reasoning which does. So when you say it's "invalid" you don't mean because of "form" because there is nothing wrong with the form, you mean because the conclusion doesn't follow from the premises. In a probablistic M.P the conclusion is a probability) and that is how Dawkin's expressed it.
Other than you don't think the conclusion follows you've not given any particular reasoning. In effect you don't accept Dawkin's premises, but that doesn't detract from his argument because he acknowledges some people won't. I think though atheists on the whole would accept the reasoning he gives. For example I do accept that if complexity and design is why the H. God is speculated then how does one explain how that God was created..it too would need a creator and creators of infinite regress. As Dawkins writes in the Blindwatchmaker, if "we had infinite time and infinite opportunities anything is possible" but the thing is if we don't have infinite time and opportunities then in theory a probability of the likelihood of life developing could be calculated if we knew the opportunities and time.
Now Sober does disagree with an upper limit probability estimate Dawkins used for probability of life in the universe in the Blindwatchmaker p 142- 146, however at this point I'm not sure that he and Dawkins are referring to the same thing. I think Dawkins was using an upper limit just as an example of a very high improbability, not that he was assuming he had reached a conclusive upper limit. In the God delusion the point Dawkins is making is that complexity doesn't spontaneously erupt..so a God couldn't spontaneously erupt..but rather any intelligence that comes about has come about through gradual processes using the physical laws of the universe.
So although you've cited Sober as somehow destroying Dawkin's argument in the God Delusion, in particular the one you quoted, Sober doesn't. Sober I believe is saying just because we exist and we know how many potential planets there are in the universe, doesn't mean we can assume a reliable upper limit probability for chances that life began based on that. But that's not Dawkins argument in The God Delusion..in that he's saying no intelligent entity could spontaneously erupt it would need to happen by evolution and happen after the existence of the universe.
This video relates though I haven't seen it yet.
http://www.youtube.com/watch?v=wa55s9Gs_Eg&feature=related