Episode 045: What if God Were Really Bad?

apologetics atheism Philosophy of Religion podcast

Here it is, the last podcast episode for 2011. This time I’m looking at “the “evil god challenge” as posed by Stephen Law in a fairly recent article by that name. Isn’t the evidence for a good God really no better or worse than the evidence that an evil god? In short, no. Here I explain why I think (as I suspect many may think) that the evil god challenges has major philosophical shortcomings, in spite of being an argument worthy of our attention.

 


Similar Posts:

If you liked this post, feel free to help support this project.

{ 97 comments… add one }

  • colin December 30, 2011, 6:52 am

    is there something missing here? I don’t see the actual audio.

  • Cameron W Gilchrist December 30, 2011, 8:46 am

    No audio link here.

  • Glenn December 30, 2011, 11:10 am

    Ah… you know what I’ll do now? I’ll actually embed the audio file now so you can listen to it!

    That was Evil-Glenn up to his usual tricks…

  • Dicky P December 30, 2011, 12:19 pm

    Evil-Dicky P

  • Cameron W Gilchrist December 31, 2011, 3:03 am

    But wait… which one of you is the real evil-glenn? Maybe both of you. Hey, isn’t that Dualism? All kidding aside thank you very much for this podcast.

  • The Atheist Missionary December 31, 2011, 12:03 pm

    The difference between Dr. Peoples when Prof. Law is present to defend himself and when Prof. Law is absent is remarkable. Given how Prof. Law’s paper is apparently such an abject philosophical disappointment, I look forward to Dr. Peoples preparing an academic paper in which he clearly sets out his objections to the evil-god challenge. About all I heard in the podcast was that “classical” theism presupposes God as being good and therefore the challenge never gets off the ground. I, like Prof. Law, am still waiting for Dr. Peoples to clearly set out the premises of his moral argument in support of the existence of a good god.

  • Glenn December 31, 2011, 12:25 pm

    TAM, when I discussed the moral argument with Stephen Law on the radio show, I said exactly the same things about the plausibility of the moral argument, exactly the same thing about the failed response to the “moral sense” argument, and exactly the same thing about the irrelevance of his appeal to the numbers of people who rejected the moral argument. The only reason I didn’t also say the same things about the Euthyphro dilemma in that discussion is that there wasn’t time to discuss that. I wasn’t able to set the agenda for what material we covered.

    So I don’t know what this “marked” difference is. I suspect you’re talking about the fact that I spoke about all the other considerable problems with the evil god challenge in this episode that I didn’t talk about on that show. But that’s perfectly understandable, as I on;y had the opportunity to discuss one argument on that show – the moral argument!

    As for the “I’m still waiting” ruse, remember this: Stephen set out in his paper to offer a challenge to theists. It’s no good to say: “Right, here’s a challenge. You guys do all the work, proving that all your beliefs are true and all your arguments work.” Sure, there’s a place for that, but it’s not a challenge to what theists actually believe.

    Also, I think you’ve misunderstood why Stephen raised that demand about the moral argument. I did offer plausible considerations for the conditional premise of the moral argument in a recent blog post. Instead of offering a single reason why we should doubt my explanation, Stephen asked me to lay it out in premises – because of his mistaken view about how probability works, and based on his view that the conclusion of every deductive argument is determined by multiplying the probability of each stand-alone premise. You evidently missed the discussion around that.

    About all I heard in the podcast was that “classical” theism presupposes God as being good and therefore the challenge never gets off the ground.

    Then you were simply not listening. That’s not even a decent summary of what I said about classical theology, let alone the other areas I discussed (which you evidently missed altogether, if the above is all you heard).

  • Kenneth December 31, 2011, 1:15 pm

    The Atheist Missionary:

    “About all I heard in the podcast was that “classical” theism presupposes God as being good and therefore the challenge never gets off the ground.”

    Wow. That’s absolutely shocking comprehension. Glenn didn’t even talk about what classical theism “presupposes.” He talked about what they believe, and noted that to understand the reasons for it you have to become familiar with a lot of background information.

    And this (misunderstood) claim is the only thing you heard? That’s simply incredible. I heard a lot more.

    * I heard Glenn explain that classical theism maintains that there is a unity of transcendentals, which means that without actually showing that its metaphysical outlook is false (and Glenn noted that Stephen didn’t do this), the evil god challenge is not a challenge.

    * I heard Glenn refer to a moral argument for theism and I heard him describe Stephen’s response to it where he just appealed to the oft-trounced appeal to the Euthryphro dilemma, which (in addition to being oft-trounced) is not even an objection to the moral argument, but is an objection to divine command ethics, which is not the same thing at all.

    * I heard Glenn describe Stephen’s summary and response to the “moral sense” argument, and I also heard Glenn utterly demolish this by explaining how badly Stephen misconstrued the moral sense, thereby revealing a major gap in Stephen’s understanding of the material he was discussing. Glenn explained how the combination of the moral argument and the moral sense argument actually does offer support for a good God over and evil god.

    * I also heard Glenn summarize and respond to Stephen’s treatment of the whole area of historical apologetics, explaining that Stephen allows the evil god believers to use a kind of argument that Stephen himself elsewhere rejects as disingenuous.

    * I also heard Glenn put his finger right on the button and identify the fact that Stephen evaded the hard philosophical work at every single step: He never explained why it’s absurd to allow the theodicies to stand. He never explained what’s actually wrong with the moral argument. He never explained what’s wrong with historical arguments. He never grappled with the issue of classical theology, and yet Stephen was so philosophically in the dark in this subject area that he repeatedly took himself to be specifically addressing “classical theism”!

    So if the only thing you heard was ‘ “classical” theism presupposes God as being good and therefore the challenge never gets off the ground,’ then you just weren’t listening – or else most of what was said just went right over your head.

  • SpiritualKiss January 1, 2012, 1:16 am

    Glenn,

    Thanks everso much for the podcast – I really enjoyed it. I could sense Stephen Law itching to get a word in edgeways, because no-one really understands the EGC unless he is there to helpfully explain it to them!

    I appreciated your analysis of what the Classical theist’s responses might be to the EGC. As you know there was a discussion over at Edward Feser’s blog a couple of months ago and so it was interesting to hear your take on it too. Back then, there were two things that Dr Law claimed: (1) that the EGC could be run while holding a privation view of evil and (2) that in any case it didn’t ‘meet’ the challenge, it attempted to avoid it.

    On the first point, I’m not sure I follow Dr Law – unless he thinks that there is an insurmountable difficulty posed by the EPOE, reinterpreted by his EGC. But as you say, the Classical Theist can dispose of the Evil God hypothesis which leaves the age-old EPOE. And so it sounds like a privation of an argument!

    The point that Stephen Law has yet to answer to my satisfaction was HOW one might rule out an evil creator on empirical grounds (in order for the EGC to continue to be run). He appeals to intuition as to how a theist might accomplish this(!?) This seemed to me to be very poor reasoning and one which Dr Law refused, upon request, to even bother to answer.

    These are the issues that concerned me – I’m sure others will have their criticisms of Dr Law’s EGC. I hope that this podcast, in addition to the Unbelievable programme, will help to hold Stephen Law to the claims he is making, since he appears happy for his challenge to remain ambiguous – and so it can actually mean anything he wants it to mean.

    Regards, SK.

    PS In your quote from Richard Dawkins, did I hear him right when he calls ‘reading innermost thoughts’ a human attribute?!

  • Dave January 1, 2012, 9:08 am

    Interesting observation for me: When it comes to deciding that the reverse theodicies fail, Law relies exclusively on intuition and this crucial reaction is what grounds the entire evil God challenge, because based on this intuition-grounded claim he wonders why theists don’t take the same view of their own theodicies.

    And yet in discussion with Dr Peoples, Law says that moral facts, being believed in only on the basis of intuition, therefore have a weak basis and could easily be jettisoned in order to escape the moral argument!

    The evil god challenge has taken a bit of a whipping!

  • Jonathan Deundian January 3, 2012, 12:41 pm

    If “good” and “evil” (evil being privation) represent an objective reality, then the explanation of that reality is, probably, God or something much like Him, NOT Platonic realism as Stephen Law suggests. Is Platonic realism supposed to be less metaphysical? :-)

    I haven’t listened to you podcast yet…, sorry for the random comment.

  • Stephen Law March 1, 2012, 12:31 pm

    Commentator above said about me: “He never explained why it’s absurd to allow the theodicies to stand.” That’s right, I didn’t do that in the paper.

    That’s because I didn’t think anyone would actually take the line that the reverse theodicies are entirely successful in dealing with the evidential problem of good. I just thought that move was too nutty and desperate even for most theists, to be honest.

    However, surprisingly, that’s the line Glenn has taken, and I did deal with it at great length in a discussion on this blog. Indeed, Glenn proved himself unable to explain why his response was any more reaaonable than that of a Young Earth Creationist to the fossil record etc. (who use the same strategy of ad hocery and mystery mongering to deal with such counter evidence). None of that gets a mention here of course.

    And, as I repeatedly pointed out in the radio interview with Glenn, if the theodicies fail, Glenn comes out as deeply irrational even if the first premise of his moral argument (that there’s no good without God) can be established (which it can’t). That’s also ignored here.

    Basically, I have established Glennis deeply irrational. Though obviously he is never going to accept that.

    PS my point about the moral argument and probabilities was basically right. If an argument is (i) deductively valid and (ii) has 5 non-redundant premises of prob 0.8 each, then the highest probability you can reasonably assign the conclusion given just those two facts alone is indeed 0.33. The highly technical stuff about upper bounds boiled down to the observation that given other factors it might be reasonable to assigna a higher probability to the conclusion. Well, sure. I knew that already. I am not sure Glenn was able to follow that discussion though (which involved a very knowledgable Tim).

    PS I shan’t comment again here. Got a paper to write…

  • Glenn March 1, 2012, 9:58 pm

    Stephen… seriously, again? You were unable to make a decent parallel between Young Earth Creationism and theodicies, as far as I could tell. I’m loathe to repeat the whole thing now, so I’ll just link back for those who are interested in seeing what was said (and they can also listen to this episode). They can see you hanging your whole response on a very hopefully repetition of “ad hoc” and decide for themselves the quality of your argument.

    As for the condescending stuff about me not being able to follow the later discussion about probability – as in, the discussion where it was explained why Stephen’s comments about probability was quite wrong – rest assured, I followed it. I don’t know why Stephen would say otherwise. The upshot was that Stephen’s assessment of the moral argument’s probabilistic strength was simply mistaken, and he was merely gratuitously assuming that the only grounds for supposing that God is good is the moral argument, AND also that the moral argument requires a long list of premises such that the final probability is low. This is mistaken, since it actually has two such premises (the other version I presented boils down to these three steps, two conclusions and a premise), which with a probability of .8 for each premise still yields an overall probability of .64 (more probable than not) – and that’s even IF we let Stephen off for his error of assuming that in a deductive argument this is how you establish the probability of the TRUTH of the conclusion.

    See Stephen, not only did I understand it, I also understand that you havent appreciated the importance of your mistake, a mistake I didn’t actually see you explicitly admit. :) You learned something, and instead of conceding that, you somehow now want to make out that it was I, and not you, who didn’t understand to begin with, in spite of the fact that what you learned is what I already knew and tried in vain to tell you!

    For what it’s worth, you’re welcome.

  • Stephen Law March 1, 2012, 10:05 pm

    “Seriously again?” Yes. If you continue to misrepresent and distort what was said like this, expect me to show up and correct you.

  • Glenn March 1, 2012, 10:10 pm

    Now I am misrepresenting your argument? Where? This is news. Sure, “stand up” if you must, but unless you have something to point out and correct…

    And as for distort – it is not a distortion to point out that you made a claim about probability in regard to the moral argument, I pointed out that this claim was false, you replied and insisted that no, your claim was really true – which I was surprised by – and then Tim chimed in and pointed out that no, your claim was really wrong – and now you reflect on the whole episode as something that I just didn’t understand. Simply marvellous!

    But I’m the one distorting what happened…. :)

  • Stephen Law March 1, 2012, 10:29 pm

    The upshot was that Stephen’s assessment of the moral argument’s probabilistic strength was simply mistaken” Why? It wasn’t mistaken – it was entirely on target. It just needed phrasing a bit more accurately.

    “,,,and he was merely gratuitously assuming that the only grounds for supposing that God is good is the moral argument”

    Doh, I never thought of that. What I fool I have been. But what’s it got to do with my assessment of the strength of the moral argument?

    “AND that the moral argument requires a long list of premises such that the final probability is low.”

    The first premise requires significant support – various things need ruling out, as you acknowledged in your informal presentation of it. So there several suppressed premises. However, when I asked you to spell them out in a formal version so we could count them and assess their probability, you refused!

    I’d call that running away…

    On my observation that ad hocery and mystery mongering being what you, Young Earth Creationist and various nutters have in common re defensive strategy, I am of course entirely correct. You first misunderstood what ad hoc meant, then tried denying you were making ad hoc moves, then resorted to insisting it didn’t matter that you were, without ever providing a decent explanation of why (except to say your moral argument saves you, but – even with an established first premises – the moral argument does significantly raise the probability of God unless you have FIRST dealt with the PoE, for the reasons I explained. The first premise just combines with the PoE to deliver the conclusion there are no objective morsl values – which was only a somewhat suspect intuition in the first place).

    You’re in an Intellectual black hole, Glenn. Just like many Young Earth Creationists, Christian Scientists, etc. – people who hold entirely different religious beliefs, yet are as convinced as you, and just as set on shoring up the belief of the faithful and gaining more True Believers, as you…

    Does that thought never, ever, cross your mind?

  • Stephen Law March 1, 2012, 10:37 pm

    oops that should have read “does NOT significantly raise the probability of God”. No doubt you’ll have that down as a Freudian slip!

    Anyway, enough of this bad tempered exchange….

    The discussion I have had with you has been very valuable. You are patine and clear when it comes to explaining stuff. And Tim was very helpful indeed in making me see I needed to express myself a bit more accurately re probability, which I then did. So thanks for all that…

  • Stephen Law March 1, 2012, 10:38 pm

    patient not patine, whatever that is. must read things before hit “send”

  • Glenn March 1, 2012, 11:09 pm

    You are welcome. I’m sure your comments are at least partly the product of frustration, so I won’t be bearing any grudges. :)

    For readers who aren’t sure what we’re talking about with this probability stuff and who think that maybe I didn’t understand what was going on – it relates to my reply to Stephen’s comment here, and the discussion that followed. Check for yourselves. Reading back through it, I am reminded of how in fact I understood quite well what was being said about probability, and yet at the end of it, I don’t think Stephen did. Far from running away, I pointed out fairly directly how Stephen’s claims about the probability of my conclusion were straightforwardly untrue – and my explanation was validated by others to whom Stephen appeared to defer. But decide for yourselves.

    In addition, they can check the previous link I gave – just to check the truth of Stephen’s assessments of what was said in regard to ad hoc arguments etc. They may find that Stephen’s confident recollections are a little embellished. :)

  • basil March 2, 2012, 5:38 am

    ‘You’re in an Intellectual black hole’

    I hear that from my professors everyday… It builds confidence.

  • Geoff March 2, 2012, 10:35 am

    It appears grace is not an “evolutionary enhancement” for atheists..

  • Glenn March 2, 2012, 7:44 pm

    Just one more comment on the whole probability thing: “Doh, I never thought of that. What I fool I have been. But what’s it got to do with my assessment of the strength of the moral argument?”

    Actually rather a lot. Remember that your attempt to assess the highest possible probability of the moral argument ended up (due to an error in calculating the probability where deductive arguments are in view) actually describing the method of establishing the lowest possible probability probability.

    It was explained to you then by Tim (yeah, I did follow it. I learned good!) that the actual probability of the conclusions truth (which you were aiming at) would need to take into account whether or not there were any other reasons for supposing that the conclusion is true apart from those included in the specific argument in question.

    That, in short, is why whether or not there are any other reasons to think that God is good is actually very important when it comes to assessing your claim about the conclusion’s probability.

  • Stephen Law March 3, 2012, 12:33 am

    Glenn. you say: “That, in short, is why whether or not there are any other reasons to think that God is good is actually very important when it comes to assessing your claim about the conclusion’s probability.”

    Yes OF COURSE it is!

    But the degree of support an argument gives for a conclusion is NOT determined by the conclusion’s overall probability. The probability of the conclusion could OBVIOUSLY be very high even if the argument is absolutely useless – please acknowledge this Glenn.

    So, your switch to talking about the overall, all-things-considered probability of the conclusion, in response to my concern about the strength of your moral argument (which was only ever raised as a concern because we did not – and still don’t – know how many suppressed premises it has or what they are), is a complete red herring.

    At this point, your claim to have a strong moral argument is absurd. BUT remember, even if you COULD establish the first premise (no moral value without God), the argument STILL fails if you have not dealt with the evidential problem of Evil, which you obviously haven’t (you say “the theodicies deal with that!” but refuse even to say which theodicies supposedly work, or why they work, and in any case, as I pointed out, they all ultimately rely on ad hocery and mystery mongering, just like the explanations of Young Earth Creationists and many nutcases).

    Doesn’t it now strike you just how remarkably weak your overall rational position is, as you have argued for and defended it? You have failed to deal with the obvious argument against it, your main argument for, in which you place so much confidence, is (i) obscure (what is it, even?!) (ii) looks prone to the worry I raised about probabilities (a worry you have not addressed but instead resorted to red herrings), (iii) in any case fails if you have not dealt with the PoE, because the PoE in combination with the first premise that there’s no moral value without good, just delivers the conclusion there’s no moral value. That there is moral value was only ever an (admittedly strong) intuition – an intuition that already looked a little shaky for other independent reasons (it’s one were likely to have even if there are no objective moral values).

    Am I not right, then, to be skeptical about your particular God claim? Should not you, at the very least, be far more cautious about it?

  • Thomas Larsen March 3, 2012, 2:35 pm

    Stephen, you’re familiar with the main theodicies suggested by orthodox Christian theists. Why don’t you accept them? (Perhaps they are developed in a fairly ad hoc fashion; perhaps their cousins, defences, make—quite justified, in my opinion—appeals to mystery; but what does any of that have to do with the usefulness of theodicies or defences?)

    Heh. I’m a fan of many-world theodicies.

  • Glenn March 3, 2012, 3:29 pm

    Stephen… look, I’ll quote exactly what you said at the end of the previous discussion about probability: You said – even after the discussion with Tim, that your multiplying of the probability of the premises gives us the “highest probability we are licensed to place on the conclusion” (from your comment 37 on this page). That was right at the end of that thread.

    So you were still saying that your multiplication gave the HIGHEST possible probability we could assign to the truth of the conclusion. But as I had earlier said (you didn’t seem to believe me), and as Tim tried to explain to you, this is wrong. What you had actually calculated was the lowest bound of the probability. The actual probability is not established by the argument in question, only the lowest bound is. And now you’ve been saying (see above), that it was I who might not have followed that discussion. It’s a bit frustrating!

    So it’s not a “switch” for me to talk about the overall, all-things-considered probability. Indeed, what you calculated was only a lower bound, so it would be a switch for you to move from that to talking about any actual probability.

  • Stephen Law March 3, 2012, 3:41 pm

    For f*ck’s sake Glenn. You just selectively quoted only part of what I said. Here’s the full quote:

    “It seems, given what Tim has said, that the highest probability we are licensed to place on the conclusion of an argument, by virtue of its being valid and having five non-redundant premises of probability of 0.8 each, is 0.32768.”

    Read it carefully. Notice the part you selectively edited out and think about what it means.

  • Glenn March 3, 2012, 4:02 pm

    Stephen, I did look at it. Now you look, patiently. Look at what I corrected – the word “highest.” You said that the highest probability that we are licensed to place on the conclusion, given that it’s valid and given those probabilities, is 0.32768.

    This is flatly wrong, as was explained in that thread. That is exactly where you went wrong, because this is not the highest probability at all, as was explained to you by Tim. What’s frustrating about it is that you claimed the opposite, and you claimed to be doing so on the basis of what Tim said!

    In fact 0.32768 is the lowest possible probability we can assign to the conclusion, given the argument’s validity and given the probability of the premises (although given my argument with two premises with 80% probability each, this should have been 0.64).

    So I was not “selectively” quoting you at all, I was quoting the salient part, as I trust you now see. Your claim was wrong, and the reverse claim is true.

    What your attempt at calculation would actually have given was the lower bound of probability. Not an actual probability, and certainly not the “highest” probability based on the premises given and their independent probability.

    Do you see that now?

  • Stephen Law March 3, 2012, 4:16 pm

    0.32… is the highest probability we are licensed to place on the conclusion given no more than that an argument that has (i) deductive validity and (ii) 5 non-redundant premises of prob 0.8 each.

    The “lower bounds” point is that the conclusion may yet have a higher probability for other reasons (e.g. other strong arguments for same conclusion, probabilistic relations between premises, etc.) as we agreed ages ago. That’s irrelevant to the question of what is the highest probability we are licensed to place on the conclusion given just (i) and (ii).

  • Glenn March 3, 2012, 4:23 pm

    I trust that you have implicitly retracted your claim about me selectively quoting you.

    But as for the substantive claim – Again – No. The valid argument with premises of a known probability tells us the lowest bound of the actual probability of the conclusion. That’s what it does. That’s all it does.

    Saying that it tells us the “highest” probability that we can assign to the conclusion on the basis of the argument is clearly not right, because that implies that the probability might in fact be lower. But it cannot possibly be any lower, because the product of the multiplication here is the absolute lowest bound. What the equation does is to tell us the lowest probability. That’s it.

    So Stephen, I would ask why you chose to say “highest” rather than “lowest,” since it was explained to you that while the actual probability might be higher, it literally could not be lower. (And let’s remember, 0.64, not 0.32). And that’s not rhetorical. Given that “lowest” really is the case as the actual probability could not be any lower, I actually am asking you why you chose to say “highest.” At the risk of sounding patronising as that’s really not my intention at all, I’d ask you to recall that initially you thought this was the way to calculate the actual probability, and you dismissed my concern as simply mistaken. Now you’ve come around on that claim, please look again at the above.

  • Stephen Law March 3, 2012, 4:33 pm

    “Saying that it tells us the “highest” probability that we can assign to the conclusion on the basis of the argument is clearly not right, because that implies that the probability might in fact be lower. ”

    Ah I see your confusion, Glenn. Actually, saying 0.32… is the highest probability that can reasonably be assigned to the conclusion given just (i) and (ii) does NOT imply the probability of the conclusion may be lower, given just (i) and (ii), does it?

    True, the probability of the conclusion, all things considered, might indeed yet turn out to be lower than 0.32… (but that would indeed require that the probability of one of the premises is then also lower than 0.8).

  • Stephen Law March 3, 2012, 4:35 pm

    Nope, you did selectively quote me. Obviously so. You missed out the “given..” part of what I said, turning a conditional claim into a non-conditional claim. That is clearly and obviously a case of selective quotation.

  • Nathan March 3, 2012, 4:39 pm

    It seems to me that that’s the crux of it: Stephen called the lowest the highest, when it isn’t, and I couldn’t find the justification for it. I’ve only looked at all these comments (and the history) today and that’s how it comes across.

  • Glenn March 3, 2012, 4:40 pm

    True, the probability of the conclusion, all things considered, might indeed yet turn out to be lower than 0.32… (but that would indeed require that the probability of one of the premises is then also lower than 0.8).

    For goodness’ sake, Stephen… This is just the same as admitting – without saying it in as many words – they in fact the probability of he conclusion given the stated probability of the premises could NOT in fact be any lower than 0.64 (I keep reminding us that this is the relevant figure). To say “Oh but really it could, if we change the conditions of my conditional claim” is just to say that things would be different if they were different. Just concede the point directly rather than beating around the bush like this.

    And no, I didn’t selectively quote you, because the fact that your statement was conditional is not the part that I was correcting you on. I was pointing out why “highest” should be replaced with “lowest.”

  • Stephen Law March 3, 2012, 4:41 pm

    Put it like this, Glenn. Given just (i) and (ii), we can reasonably assign no higher or lower probability to the conclusion than 0.32, though of course the probability of the conclusion may yet turn out to be much higher, or, indeed, lower (though the latter would requires the rejection of (ii))

  • Glenn March 3, 2012, 4:42 pm

    Again Stephen…. (ii) is one of the conditions that you stipulated!

  • Stephen Law March 3, 2012, 4:43 pm

    “And no, I didn’t selectively quote you, because the fact that your statement was conditional is not the part that I was correcting you on. I was pointing out why “highest” should be replaced with “lowest.””

    But that “lowest” is the right word to use only follows if my claim is understood non-conditionally! That’s exactly my point.

  • Glenn March 3, 2012, 4:43 pm

    Obviously we’re typing at the same time…

    The point is that given i) and ii) (correcting it to a two premise argument), we can’t deny that the probability of the conclusion is higher, but we can definitely rule out that it’s lower.

    (And re: your last comment, “lowest” is the case even given your conditional claim. So if that was your point, your point is mistaken.)

  • Stephen Law March 3, 2012, 4:48 pm

    “The point is that given i) and ii) (correcting it to a two premise argument), we can’t deny that the probability of the conclusion is higher, but we can definitely rule out that it’s lower.”

    Yes, indeed, we can’t deny that the all-things-considered probability of the conclusion might be higher. I have said that many, many times. That’s obvious. My point is WE ARE NOT JUSTIFIED IN GIVING THE CONCLUSION A HIGHER PROBABILITY THAN 0.32.. ON THE BASIS OF JUST (I) AND (II) ABOVE. That’s my point.

    Given just (i) and (ii) above, the HIGHEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…

  • Stephen Law March 3, 2012, 5:00 pm

    ps I am talking about a hypothetical argument about which we know no more than (i) and (i). I don’t claim it’s your moral argument. I have no idea how many premises your moral argument has when all the suppressed premises are set out. How can I know, when you refuse to say what they are?!

  • Glenn March 3, 2012, 5:02 pm

    Stephen, are you willing to say with equal force and repetition:

    “Given just (i) and (ii) above, the LOWEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…”

    (recognising that you’re talking about five premises and I’m talking about two, so the figure would be 0.64)

    If not, why not? It looks to me like this choice has been simply made because of partisanship.

    And my last comment on this – I see that you keep talking about assigning *a* probability based on this two premise argument (although you’re talking about a five premise argument). Something I’ve said already is that on the basis of this argument we’re actually not justified in assigning a specific probability at all. We’re only justified in establishing a boundary – a lowest bound. If we’re being careful and not overstepping what we’re justified in saying, the MOST we can say is “whatever the actual probability is, it can only be this much or higher.”

    We are doing things wrongheadedly if we end up saying: “On the basis of this argument, the probability is no HIGHER than….”

  • Stephen Law March 3, 2012, 5:05 pm

    ((Stephen, are you willing to say with equal force and repetition:“Given just (i) and (ii) above, the LOWEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…”))

    Yes! For the final time, shouting it from the mountain tops, abso-frigging-lutely!

    And are YOU, Glenn, now willing to say:

    “Given just (i) and (ii) above, that the HIGHEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…”

    ??

  • Stephen Law March 3, 2012, 5:09 pm

    Also, re you also willing to concede that the all-things-considered probability of the conclusion may be lower than 0.32 (though of course that would also require downward revision of the probability of at least one premise).

  • Stephen Law March 3, 2012, 5:11 pm

    PS what I mean by that is that our initial assessment of the probabalities of the premises (as being 0.8) may yet need to be revised downwards in the light of other evidence and arguments…

  • Glenn March 3, 2012, 5:14 pm

    “Yes! For the final time, shouting it from the mountain tops, abso-frigging-lutely!”

    If you had said it once so far, that would have been good!

    And are YOU, Glenn, now willing to say:

    “Given just (i) and (ii) above, that the HIGHEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…”

    As I have said, we are not justified in assigning any probability. I have tried to say this a few times now. We can’t do it – not, at least, on the basis of a two premise deductive argument like this one. This is in pat why I think it’s misleading to you saying that we can’t “assign” a probability higher than 0.34 (should be 0.64), as though we would be justified in assigning that probability. We wouldn’t.

    It’s literally impossible, Stephen! So it’s a bit of a trick question. What’s the highest probability we can assign? Well it’s not 0.64 (the lower bound) – or anything else. We can’t assign on this basis!

    All we can do – absolutely all we can do, on the basis of the two premise deductive argument where each premise has a probability of 0.8, is establish a bottom limit below which we know the actual probability is not. If we do more on the basis of this two premise argument, we overstep our entitlement.

    “Also, re you also willing to concede that the all-things-considered probability of the conclusion may be lower than 0.32 (though of course that would also require downward revision of the probability of at least one premise).”

    Well given i) and ii), neither of us accepts that the all-things-considered probability could possibly be lower. Not you, and not me. If we throw out i) and ii) and decide that the probability of the premises is lower and we also decide that there are no other reasons to accept the conclusion than the premises in this argument (which would be very generous of me), then yes, the probability of the conclusion might be lower. But given the conditions you stipulated, that’s literally impossible.

  • Stephen Law March 3, 2012, 5:30 pm

    Glenn’s comment at 44 (regarding my replying: Yes! And for the final time, abso-frigging-lutely to: “Given just (i) and (ii) above, the LOWEST probability we’re justified in assigning the conclusion of that hypothetical argument is 0.32…”):

    “If you had said it once so far, that would have been good!”

    Stephen’s omment at 34. “Given just (i) and (ii), we can reasonably assign no higher or lower probability to the conclusion than 0.32″.

  • Glenn March 3, 2012, 5:44 pm

    Not the same.

    You included “higher” as well as lower in that statement. I was just noting that you repeatedly said the highest probability we can assign is X, giving great emphasis to this claim without reference to the lowest probability, but not once made the parallel statement: “The lowest probability we can assign is X” – until you were directly asked. And even then, all you could muster was “yes.”

    But it looks like all the dregs of this are mopped up now. The actual probability of the conclusion of the moral argument (granting a probability of 0.8 for each premise) is at least as high and perhaps higher than 0.64 – but definitely not lower. It will be higher if there are other reasons for thinking it to be true, but nothing can make it lower. Lower probabilities are off-limits. Higher probabilities are possible. So it is between 0.64 and 1.

    Strong language and terse replies aside, that’s where things sit (granting that the probability of 0.8 for each premise is at this stage arbitrary). This has panned out pretty much as I thought it would.

  • Stephen Law March 3, 2012, 5:45 pm

    Glenn, you say. “All we can do – absolutely all we can do, on the basis of the two premise deductive argument where each premise has a probability of 0.8, is establish a bottom limit below which we know the actual probability is not.”

    Right. So you agree that we are not entitled, on the basis of my (i) and (ii), to assign a higher probability to the conclusion of that hypothetical argument than 0.32…

    But that was my point all along! Here is the exact thing I said above, which you objected to:

    “It seems, given what Tim has said, that the highest probability we are licensed to place on the conclusion of an argument, by virtue of its being valid and having five non-redundant premises of probability of 0.8 each, is 0.32768.”

  • Stephen Law March 3, 2012, 5:50 pm

    Glenn, you say:

    “Not the same.

    You included “higher” as well as lower in that statement.”

    Are you serious? Suppose I say I love cheese and I love chocolate, and someone says, but do you like cheese? I say: yes, I just said so. Would I be right? If the reply came; No you said you love cheese AND you love chocolate, surely we could only reasonably conclude that we were dealing with either an idiot, or an intellectual trickster and charlatan.

  • Glenn March 3, 2012, 5:50 pm

    “Right. So you agree that we are not entitled, on the basis of my (i) and (ii), to assign a higher probability to the conclusion of that hypothetical argument than 0.32…”

    Once more: we are not justified in assigning any probability here. None. Nothing at all. The only thing we can say is that the actual probability cannot be lower than 0.32. That is not what I quoted you as saying at all.

    “Definitely cannot be lower than” is not the same as “cannot justifiably be said to be higher than.” I quoted you saying the former, not the latter.

  • Stephen Law March 3, 2012, 5:54 pm

    “Once more: we are not justified in assigning any probability here.”

    In which case I am indeed correct. We are not justified in assigning a probability higher than 0.32…

  • Kenneth March 3, 2012, 5:57 pm

    But it looks like all the dregs of this are mopped up now. The actual probability of the conclusion of the moral argument (granting a probability of 0.8 for each premise) is at least as high and perhaps higher than 0.64 – but definitely not lower. It will be higher if there are other reasons for thinking it to be true, but nothing can make it lower. Lower probabilities are off-limits. Higher probabilities are possible. So it is between 0.64 and 1.

    Strong language and terse replies aside, that’s where things sit (granting that the probability of 0.8 for each premise is at this stage arbitrary). This has panned out pretty much as I thought it would.

    Game > set > match

  • Glenn March 3, 2012, 5:58 pm

    In which case I am indeed correct.

    LOL…. good night. You’re on an iceberg that has melted to become an ice cube, and you’re now celebrating the victory that you’re still afloat?

  • Thomas Larsen March 3, 2012, 5:59 pm

    That last comment of mine apparently didn’t submit properly (Glenn, please delete it)…

    Stephen, your claim seems to be that, given an argument of the form

    (1) Premise 1
    (2) Premise 2
    (3) Conclusion,

    Pr(Conclusion) <= Pr(Premise 1) * Pr(Premise 2).

    I think that Glenn, however, is saying,

    Pr(Conclusion) >= Pr(Premise 1) * Pr(Premise 2).

    Correct me if I’m wrong.

  • Glenn March 3, 2012, 6:01 pm

    Deleted as requested, Thomas.

  • Stephen Law March 3, 2012, 6:17 pm

    Thus if we know about some hypothetical moral argument only that (i) it has five premises (ii) of prob 0.8 each, we are not thereby justified in assigning a higher probability to its conclusion than 0.32.

    Hardly a powerful argument, as it stands.

    Is your argument such an argument? We don’t know, because you refuse to say what it is (with supporting premises spelled out).

  • Stephen Law March 3, 2012, 6:20 pm

    No Thomas I’m not claiming that.

  • Glenn March 3, 2012, 6:23 pm

    Stephen, your claim seems to be that, given an argument of the form

    (1) Premise 1
    (2) Premise 2
    (3) Conclusion,

    Pr(Conclusion) = Pr(Premise 1) * Pr(Premise 2).

    I think that Glenn, however, is saying,

    Pr(Conclusion) >= Pr(Premise 1) * Pr(Premise 2).

    Correct me if I’m wrong.

    I’m saying the latter.

    And as I’m sure Stephen won’t mind me speaking for him… ;) (and he has now said that he’s not saying the former.)

    Stephen’s not saying the former. When he is asked directly for an answer, he much less enthusiastically concedes that the latter is the case.

    The difference between what Stephen and I are saying is that what I am saying is what follows from the probability calculation (where Pr(Conclusion) >= Pr(Premise 1) * Pr(Premise 2)). What follows is a claim that is limited to only speaking about the lower bound of probability.

    But the lower bound is not what Stephen wants to talk about, so he’s trying to hammer a square peg into a round hole. He’s taking a calculation which actually only speaks about the lower bound, and he’s trying to apply it in some way to support his hopes about the actual probability of the conclusion. I have challenged him on this, and I’ve noted a change in his language in response to this challenge. Initially he phrased his claims in terms of “the highest probability we CAN assign” to the conclusion (my emphasis). But the calculation has nothing to say about what we should be assigning in terms of the highest probability. That’s not the function of the calculation.

    So now he has dialled his claim back to a negative one: We can’t actually, on the basis of this argument, assign any specific probability that’s higher than the lower bound. This softer claim is trivially true, only because we can’t assign at all. That’s why I commented to Stephen that it’s a bit like celebrating that he thinks he’s still afloat, even though his argument that he thought was an iceberg has melted right down to a tiny ice cube. But what we’ve established in this back n forth is that 0.62 is the lowest possible probability (of my argument), and the actual probability may be higher, we’re just not in a position – on the basis of this one argument – to know where it should be placed. Without saying so as explicitly as I have said it (because that sounds like it concedes too much), Stephen has agreed to this.

  • Glenn March 3, 2012, 6:27 pm

    “Is your argument such an argument? We don’t know, because you refuse to say what it is (with supporting premises spelled out).”

    Stephen, remember that I linked back to my formulation of the moral argument? Remember that back n forth about the fact that it had five steps as laid out, but was actually only a two premise argument? And that’s the point at which you started commenting on the probability of the conclusion of a “five premise” argument.

    And in spite of this you keep telling people that I’m too scared to spell it out.

    To reiterate, my argument is a two premise argument with a conclusion.

  • Thomas Larsen March 3, 2012, 6:27 pm

    Stephen, out of interest, what is your claim then?

    Thus if we know about some hypothetical moral argument only that (i) it has five premises (ii) of prob 0.8 each, we are not thereby justified in assigning a higher probability to its conclusion than 0.32.

    You seem to be suggesting that Pr(Conclusion) <= Pr(Premise 1) * Pr(Premise 2) * Pr(Premise 3) * Pr(Premise 4) * Pr(Premise 5)—or, at the very least, that we could only be justified in assigning such probabilities. But perhaps I’m wrong. Clarification would be appreciated.

    See Glenn’s article “A Simple Explanation of the Moral Argument” for his five-step argument.

  • Stephen Law March 3, 2012, 6:35 pm

    Indeed they can read, Glenn.

    BTW I didn’t choose 5 premises because your argument had five steps. I just picked five because your argument had four, and as you said about them: “Now obviously a number of these premises need to be defended” That’s to say, a number of premises look suspect and require supporting argument, i.e. more premises…

  • Thomas Larsen March 3, 2012, 6:45 pm

    Some random thoughts—

    Take the following argument:

    (1) Socrates is a man.
    (2) Every man is imperfect.
    (3) Therefore, Socrates is imperfect.

    Let’s say that Pr(1) = 0.99 (perhaps Socrates never existed, perhaps Socrates was really a woman and there was some massive conspiracy to cover it up, perhaps Socrates was an alien from another planet who changed into the shape of a man).

    Now Pr(2), on the face of it, seems to be 1. All men are imperfect, right? Well, not so fast! If Christianity is true, there is at least one man who is perfect, namely Jesus; and let us suppose, for the sake of simplicity, that if Christianity is not true, there is no perfect person. So suppose we assign Pr(Christianity is true) = 0.5—I obviously think it’s a lot higher, but 0.5 will do. Now Pr(2) = 0.5, does it not?

    But that means that the probability of the conclusion Pr(3) = 0.99 * 0.5 = 0.495. So Pr(\3) = 0.505; things look pretty good for Socrates.

    Background knowledge is important.

    I’m dabbling in things way out of my depth here, so maybe I’m wrong. Probability isn’t my strong point…

  • Geoff March 3, 2012, 7:20 pm

    “I’m dabbling in things way out of my depth here, so maybe I’m wrong. Probability isn’t my strong point…”

    Neither is theology?

  • Thomas Larsen March 3, 2012, 8:31 pm

    Geoff, by all means do elaborate!

  • Glenn March 3, 2012, 8:44 pm

    Stephen, when I say of my premises that they “need to be defended,” I never meant that they “look suspect.” I just meant that they are not already accepted by everyone. Just saying.

  • Geoff March 3, 2012, 11:43 pm

    Thomas,

    Jesus was human, just like you and me. I realise that you were just making an example, but I expect that you thought it had a point to make also.

    Jesus was perfect in respect to his relationship with God, but still human, and flawed (able to be tempted etc), just as you and I are.

  • david winter March 4, 2012, 1:52 pm

    Thomas,

    Your example about sums up this odd little dispute:

    You’re right that, given the assumptions you make, the lower-bound of the probability of the conclusion is simply the product of each probability. So, if you told me you had an argument that moves the bounds of a probability from 0-1 to 0.495-1 should I be impressed? I don’t really see why.

    Of course we have other reasons to think that Socrates was ‘imperfect’, and this argument is certainly not the best one you could come up with to arive at that conclusion. But the fact other arguments provide better support for a conclusion can hardly be used to assess how strong a given argument is!

    I agree with you that background information is important, and what we really ought to be doing is using Bayes Factors to measure the strength of a new argument – unfortunately that lands us in subjective terroritory and I doubt anyone would agree on just what the background is.

  • Glenn March 4, 2012, 2:38 pm

    “So, if you told me you had an argument that moves the bounds of a probability from 0-1 to 0.495-1 should I be impressed? I don’t really see why.”

    Well, given that anything above 0.5 counts as “more probable than not,” I think you should be impressed under such circumstances, David.

  • Stephen Law March 5, 2012, 2:08 am

    “To reiterate, my argument is a two premise argument with a conclusion.”

    In that ridiculously simple form it’s useless (though it’s a good rhetorical device that goes down well with the faithful). It only has power once you begin to build a case for the premise that if there is no god, there are no moral facts, and that takes more premises. You had a stab here

    http://www.beretta-online.com/wordpress/2011/the-conditional-premise-of-the-moral-argument/

    with three additional premises, but that additional argument led only to the weaker:

    “Therefore, if there are moral facts, the most plausible way to think of their basis is in terms of a supernatural person who brings moral facts about.”

    And in any case you yourself acknowledged some of these additional premises in turn needed defending. E.g. “The basis of moral facts is not natural”? Why I accept that? Yet more premises are required for support.

    Over here

    http://www.beretta-online.com/wordpress/2011/the-conditional-premise-of-the-moral-argument/

    is your attempt to argue for your conditional premise. You produce a rambling 5,000 word essay with rather hedged conclusions: “Firstly, the nature of moral facts compared to the nature of natural facts in general suggests important differences that make moral facts difficult to portray as simply features of the natural world. Secondly, the attempts by atheist philosophers to provide a natural foundation for moral facts does not look promising at all, contributing to the suspicion that if atheism’s best cannot provide such a foundation, it may simply not be there”

    Notice that even after your extended 5k word essay you are still hedging with “suggests” “does not look promising” etc. Also notice that the arguments vaguely suggested in this 5k word essay are clearly arguments involving several premises. However, it is not clear what those premises are, or exactly how the arguments run. I asked you to set the premises out clearly and you did indeed refuse.

    So, your moral argument is a two premise argument? Yeh right. So is this:

    There are no moral facts if there’s a God.
    There are moral facts.
    There is no God.

    Once you introduce the additional premises required to support your conditional premise, it’s clear your argument is going to involve several more premises. Even if there are just five basic premises with a pretty high probability of 0.8 each (and this assumption involves me being pretty generous), we won’t be justified in assigning a higher probability to the conclusion that 0.32… knowing just that (i) the argument is valid and (ii) there are 5 non-redundant premises with prob 0.8 each.

  • Glenn March 5, 2012, 7:57 am

    Stephen, many people use the word “suggests” because they are not taken by the myth that they have produced something that any rational person will be blown away with. It’s a self effacing way of writing. You should try it some time. :)

    I think the conditional premise is very plausible indeed. You say you don’t – but then I think that’s motivated. I’d love to see the list of premises required to bolster the claim that moral claims are natural! In fact – I have.

    One more thing for the moment: Stephen, I realise you’d prefer to multiply premises as much as possible in order to move the lower bound of probability down as much as possible. However, consider the following.

    The first three steps of the simple presentation of the moral argument that Thomas linked to go like this:

    1) Either the basis of morality is natural or the basis of morality is non-natural/supernatural (where these two options are to be construed as mutually exclusive).
    2) The basis of morality is not natural
    3) Therefore the basis of morality is non-natural/supernatural

    These count as three separate premises in the longer argument, so we should multiply them all and assigned a probability of .8 to them all right? (and yes, 0.8 is the figure that we’re using arbitrarily)

    Actually, no.

    Premise 1) here is analytically true, and therefore has a probability of 1.
    Say premise 2) has a probability of 0.8

    Now, step 3) is the logical deduction of 1) and 2). Therefore it has a lower bound of probability at 0.8, and we can now in fact remove steps 1) and 2) from the longer argument completely and just have 3) by itself. The above is really an independent argument for 3), and once we’ve established 3) with a lower bounded probability of 0.8, we can then take 3) as well as its lower bound of probability and use it in other arguments – such as the remainder of the simple presentation of the moral argument.

    So while I understand the desire to construe the argument as having as many premises as possible, that desire needs to be tempered with a bit of patience and bipartisanship. I have just shown why the argument that I am using is a two premise argument, and continuing to set my argument aside and instead talk about some other hypothetical argument with four premises really isn’t that relevant.

  • Stephen Law March 5, 2012, 11:31 am

    You want me to show patience and bipartisanship? Are you exhibiting patience and bipartisanship when you still:

    Fail to acknowledge I was RIGHT to say that fact that a hypothetical moral argument has five non-redundant premises of of five premises, does not justify us in assigning a higher probability to its conclusion than 0.32… to its conclusion (whether or not we’re justified in assigning any probability at all to that conclusion on just that basis [though of course we are: a conditional one])

    Fail to acknowledge that I was RIGHT to say that such an argument is, as it stands, pretty weak.

    Fail to acknowledge that I was RIGHT when I said I had made this point prior to your saying I hadn’t.

    Fail to acknowledge that your way of avoiding that last point was to say that asserting A and B isn’t the same as asserting A – either an idiotic comment or downright dishonest intellectual trickery/smokescreen from you.

    Fail to acknowledge that your conditional premise DOES indeed require supporting premises.

    Fail to acknowledge I was RIGHT to say you were refusing to unpack what those supporting premises were…

    Fail to acknowledge that the “upper bounds” point was almost entirely smokescreen, given I was never denying that the conclusion of such a hypothetical moral argument might be higher given factors other than that it has five non-redundant premises of prob 0.8 each.

    I have at least been gracious enough to admit that my original way of raising my worry was not good, and needed tightening up, which the whole upper bounds worry did allow me to do (largely thanks to the patience of Tim, not at all to you).

    You have not been gracious enough to admit one single error. You have, at every single step, dug in, using ploys and arguments that in some cases it’s hard to believe that even you didn’t know they were just smokescreen moves.

    Exposing all these errors has taken me getting on for a hundred hours of slow, painstaking analysis, over several moths of interaction with you.

    So forgive me if I decline any lectures in patience and bipartisanship from you.

    I am not sure what you do is what Jesus would do.

  • The Atheist Missionary March 5, 2012, 12:05 pm

    …… pass the popcorn.

  • matt March 5, 2012, 1:53 pm

    The jury is still out on whether or not Glenns made any important errors. Dr Law, it is plain to this reader, at least, that you are in a bad way and being very unprofessional about it. You chose to put these long hours and months into this. I think it has been an interesting exchange when you haven’t been completely childish, and I want to thank you for that. Through your Exchange with Glenn, you two have provided access to important avenues of knowledge to laypeople like myself who are very interested and can benefit from this stuff. Stop whining, sir.

  • John Griffith March 5, 2012, 2:23 pm

    This whole discussion, and not just this discussion about probabilities, is an example of why philosophy is seen as the ugly stepchild of science. It is truly a waste of your intellects. As fascinating as these exchanges might be, the whole episode boils down to “I’m right and you’re wrong”; “No, I’m right, and you’re wrong.” Can we please do productive philosophy that has even a snowball’s chance of producing something of use to humankind? Please?

  • The Atheist Missionary March 5, 2012, 2:43 pm

    John, philosophy gave birth to science. Philosophy of religion is fascinating and will provide something of use to mankind if it enables people to identify snake oil salesmen who purport to possess metaphysical certainty (ie. “there is life after death and this is how to achieve it”).

    If there is an ugly step-child, it’s theology and it’s parents are ignorance and fear.

  • Stephen Law March 5, 2012, 4:53 pm

    PS “I was RIGHT to say that fact that a hypothetical moral argument has five non-redundant premises of of five premises” should have read “I was RIGHT to say that fact that a hypothetical moral argument has five non-redundant premises of prob 0.8 each”

  • Glenn March 5, 2012, 6:13 pm

    Stephen, I don’t even know why you’re talking about a five premise moral argument at all (if not because you mistakenly thought the “five step” argument had five premises)). That’s what I’ve just explained to you in comment #25 – the same one where I suggested that you weren’t patiently reading me. I use no such argument. How apropos!

    Fail to acknowledge that I was RIGHT to say that such an argument is, as it stands, pretty weak.

    Asking me to admit – on pain of being partisan – that my argument “is pretty weak” is about as gratuitous as things get! How is that not like asking you to admit that the Evil God challenge is not a challenge at all? Or if this isn’t my argument that you’re referring to – who cares?

    You can claim that I never unpacked my defence of the conditional premise of the moral argument, in spite of complaining about how long that defence is (!!!), but just insisting that I am in the wrong unless I crumble and acknowledge that I didn’t do it… How can you not see that this is a partisan approach? We disagree on this. It’s not that I have simply refused to give up ground that I know full well I should have.

    Look, I’ll stop there. I don’t want to go on about it and sound like a nag (although you may well think I already have, and do). Your list of things that I won’t “acknowledge” is just a list of things over which we actually disagree. With respect, this is partisanship – not viewing the way the discussion went (or the way the arguments ended up succeeding or failing) in a fair way, the very thing that I was referring to. It amounts to just demanding that I acknowledge that I am wrong, and you are right.

    Now, how about that thing that, as far as I can tell, you haven’t acknowledged: Your worry about the probability of my conclusion was not mistaken because you failed to “tighten” up the way you expressed it. It was absolutely mistaken in the type of claim that you thought was justified. Let me remind you:

    In case it’s not clear, I am pointing out that a deductively valid moral argument based on even say five basic premises with an 80% probability of truth each, produces a conclusion that has 68% probability of being false. Much more likely to be false than true!

    This was not a “tightening” issue, Stephen. You’ve never acknowledged this, but I think you know that this was flatly wrong. As a philosophy lecturer, you would be right to but a big red X next to a student’s answer that went like this and to tell the student that they just haven’t understood the way deduction and probability work together. You did not understand how probability worked here. I told you, you denied it. Tim explained it to you, you accepted it (apparently) without saying “Oh, so I guess my claim was dead wrong then, and Glenn was correct.” You claimed that your multiplication trick shows that the conclusion has a 68% probability of being false. And now, if I may be so bold, you’re still riding the wave of confidence that you got from saying this, even though it’s just wrong and you’ve had to stop saying it. Now it looks like you’re trying to avoid admitting a failure to understand, saying oh well it was close, you just worded it poorly. No. granted, you’re sorted that out now, but if you want to run through a list of who won’t admit that they were wrong…

    And now I think you’re in the position of having to acknowledge that theodicies aren’t ad hoc (as per my last blog entry). Those two planks upon which you were resting your whole argument, in the end. Gone. What’s left? Not a lot, as I see things.

  • Kenneth March 5, 2012, 6:38 pm

    Glenn, you just don’t get it.

    The problem, and how we can see you’re so partisan, is that you have the audacity to think that Stephen’s not right. Only a partisan person thinks this. So what if he hasn’t given a good argument? Just admit it!

  • Sandra March 5, 2012, 7:58 pm

    Glenn, reading through this exchange (in several threads) and observing Stephen running out of room on his ever-shrinking island (or iceberg as you put it – perfect analogy!) reminds me yet again: Sometimes the best way of refuting someone is to just let them speak.

    The remarkable thing here is that Stephen is the professional philosopher who (possibly!) regards you as just a hack with a blog. And yet here it looks the other way around!

  • Nathan March 6, 2012, 12:09 am

    Stephen, obviously you feel like you’ve ultimately been successful in this argument but perhaps you need to explain yourself better, because if you are correct, the reasons why simply aren’t coming across.

    I’m sure there are many who favour you in this exchange because you represent similar views against theism, but apart from the disingenuous who would vote you the winner regardless, I’m sure most would say Glenn has the upper hand.

    I’d hate to see the conversation end prematurely.

  • Stephen Law March 6, 2012, 1:23 am

    That is a shameful response.

    Glenn you say: “Stephen, I don’t even know why you’re talking about a five premise moral argument at all (if not because you mistakenly thought the “five step” argument had five premises)). “

    It’s a HYPOTHETICAL argument Glenn, not necessarily your argument. Do you remember me saying this clearly at, for example, comment 39 (first lot comments) above, where I say:

    “ps I am talking about a hypothetical argument about which we know no more than (i) and (i). I don’t claim it’s your moral argument.”
    So I WAS right!

    P.S. Nathan you say I am not being clear enough. Is this clear enough, now? Is it clear enough now that Glenn has just misrepresented me by insisting that the points I was asking were about his argument (whatever it is), not just some hypothetical argument? Ask yourself, Nathan, whether we might reasonably suppose that Glenn actually KNOWS he just misrepresented me in order to get himself out of jail. If he does know, his behaviour is shameful. If he doesn’t know, he’s being pretty thick.

    Glenn. Please NOW acknowledge that such a HYPOTHETICAL argument is indeed pretty weak!

    Glenn, you say: “You can claim that I never unpacked my defence of the conditional premise of the moral argument, in spite of complaining about how long that defence is (!!!),”

    Misrepresentation. I complained that you never set out the supporting argument clearly in such a way that we could identify exactly what the premises were and how many there are. I was RIGHT to say you refused to do that. You ran away.

    Glenn you then say” “Now, how about that thing that, as far as I can tell, you haven’t acknowledged: Your worry about the probability of my conclusion was not mistaken because you failed to “tighten” up the way you expressed it. It was absolutely mistaken in the type of claim that you thought was justified. Let me remind you:
    In case it’s not clear, I am pointing out that a deductively valid moral argument based on even say five basic premises with an 80% probability of truth each, produces a conclusion that has 68% probability of being false. Much more likely to be false than true! This was not a “tightening” issue, Stephen. You’ve never acknowledged this, but I think you know that this was flatly wrong.”

    Yes, this is indeed flatly wrong, Glenn, if the probability is supposed to be that finally assigned to the conclusion all things considered. Happy to admit that. and in fact I admitted it ages ago. Several times. But OF COURSE I never meant the all things considered probability of the conclusion, as I have explained a million times.. So what WAS I getting at? I was getting at the point that given no more than that an argument has 5 true premises of prob 0.8 each is 0.32, we are not justified in assigning a higher prob to the conclusion than 0.32… That’s true! Isn’t it! Why not admit it? And it’s true that claim that has a 0.32 probability of being true has a 0.68 probability of being false.

    Hope this is all clear enough for you now, Nathan.

  • Stephen Law March 6, 2012, 1:29 am

    Glenn, so having dealt with yet another wave of your misrepresentations and smokescreens, let me again give you an opportunity to acknowledge your errors. Here they are, again. You still:

    Fail to acknowledge I was RIGHT to say that fact that a hypothetical moral argument has five non-redundant premises of of five premises, does not justify us in assigning a higher probability to its conclusion than 0.32… to its conclusion (whether or not we’re justified in assigning any probability at all to that conclusion on just that basis [though of course we are: a conditional one])

    Fail to acknowledge that I was RIGHT to say that such an argument is, as it stands, pretty weak.

    Fail to acknowledge that I was RIGHT when I said I had made this point prior to your saying I hadn’t.

    Fail to acknowledge that your way of avoiding that last point was to say that asserting A and B isn’t the same as asserting A – either an idiotic comment or downright dishonest intellectual trickery/smokescreen from you.

    Fail to acknowledge that your conditional premise DOES indeed require supporting premises.

    Fail to acknowledge I was RIGHT to say you were refusing to unpack what those supporting premises were…

    Fail to acknowledge that the “upper bounds” point was almost entirely smokescreen, given I was never denying that the conclusion of such a hypothetical moral argument might be higher given factors other than that it has five non-redundant premises of prob 0.8 each.

  • Stephen Law March 6, 2012, 1:36 am

    PS in comment 36 where I say “Is it clear enough now that Glenn has just misrepresented me by insisting that the points I was asking were about his argument (whatever it is)” “asking” should be “making”.

  • matt March 6, 2012, 9:02 am

    Glenn, please acknowledge that an argument you NEVER made was weak, and that Stephen is right about a trivial claim.

    Stephen, you say “Fail to acknowledge that your conditional premise DOES indeed require supporting premises.”

    but over here (http://stephenlaw.blogspot.com/2011/12/glenn-peoples-moral-argument-for-god.html) you say “Of course, Glenn realizes his premises, especially the first premise, will require considerable support, so he makes his case for it here.” and “here” is a link to:
    http://www.rightreason.org/2011/the-conditional-premise-of-the-moral-argument/#comment-13416

    So you acknowledge in one place that Glenn has offerred support for the first premise, but apparently when it is convenient you will hurl accusations at him of the very opposite thing.

    also, you say “I was getting at the point that given no more than that an argument has 5 true premises of prob 0.8 each is 0.32, we are not justified in assigning a higher prob to the conclusion than 0.32… That’s true! Isn’t it! Why not admit it? And it’s true that claim that has a 0.32 probability of being true has a 0.68 probability of being false.”

    the .32 is not the probability of the argument being true. There’s not a range of truthiness involved in the premises. They’re either true or not. Even if your probability is .32, but the premises themselves are actually true, then the conclusion is true. Going on about probabilities is, I think, simply your way of saying that you aren’t compelled to believe the argument, but that would apply to the EGC as well. People who don’t like the conclusion of the EGC can just point out reasons why the argument does not compel assent.

    Also, with respect to the falsity of the conclusion, why would we not multiply the 20% chance of the falsity of each of the premises to get to the lower bound of the falsity of the conclusion? That’s actually a real question on my part, not a rhetorical one, since I’m no expert with respect to probabilities.

  • Matthew Flannagan March 6, 2012, 10:38 am

    Stephen,
    Doesn’t your criticism prove to much: It seems to suggest that any argument which has (a) three or more premises (b) the premises are not known with certainty and (c) require additional defence is going to be a weak argument.

    That would seem to rule out almost any argument offered for any substantive philosophical conclusion. How many arguments meet conditions (a) (b) and (c), most i would think.

    Consider the following.

    [1] If God does not exist gratuitous evil does not exist
    [2] Gratious evil exists
    Therefore
    [3] God does not exist.

    Now [2] is obviously going to require some defending, one is going to have to challenge numerous premises of numerous theodicies, one is also going to have to challenge various forms of sceptical theism responses provided by various thinkers and so on. In doing so your going to need to offer a whole lot of other arguments.

    Now lets assume you offer five other arguments, let’s assume they have premises which are 80% certain. Then, by your own reasoning the problem of evil fails. Yet you defend the problem of evil in your writings.
    Or consider

    [1] If there is compelling reason for thinking God exists then there is a compelling reason for thinking and evil God exists
    [2] There is no compelling reason for thinking an evil God exists
    Therefore
    [3] There is no compelling reason for thinking God exists

    Now none of these premises, is certain lets assume [1] and [2] have a probability of 80%, premise [1] however needs defending, that involves looking at all the different arguments and reasons given for thinking God exists, different cosmological, teleological, moral arguments, ontological arguments, arguments from experience and so on, so presumably there will be half a dozen or so arguments to support [1], lets say each of these has a premise with a probability of 80% then when they are multiplied the evil God challenge becomes really weak.

  • J Smith March 6, 2012, 11:26 am

    Stephen,
    You have lost this one and it’s obvious you are wriggling furiously. Just admit it and regain some respect.

  • John Griffith March 6, 2012, 3:11 pm

    TAM, we already know how to spot a snake oil salesman. When a claim is made, ask for the evidence. If the response is anything but “here’s the evidence”, including the very popular “I don’t have any evidence, but I do have this really great argument”, you have yourself a salesman.

  • stephen law March 6, 2012, 6:00 pm

    Matt you say:

    Stephen, you say “Fail to acknowledge that your conditional premise DOES indeed require supporting premises.”

    but over here (http://stephenlaw.blogspot.com/2011/12/glenn-peoples-moral-argument-for-god.html) you say “Of course, Glenn realizes his premises, especially the first premise, will require considerable support, so he makes his case for it here.” and “here” is a link to:
    http://www.rightreason.org/2011/the-conditional-premise-of-the-moral-argument/#comment-13416

    Stephen replies. Sure I just said above that Glenn sets out a case for the premises, didn’t I? However, above Glenn backtracks to saying the argument only has two premises, because he finally realizes he has lost the battle on my point re the hypothetical 5 premise argument and realizes he is going to be in deep trouble if he doesn’t reduce the number of premises. He does this at comment 9 above, where he says:

    “Remember that back n forth about the fact that it had five steps as laid out, but was actually only a two premise argument?”

    Moreover, even when Glenn tried to make a case for the conditional premise, Glenn never set out the actual premises so we could see what they were. I asked him to. He refused. We all know why….

    Matt – I am not dealing with any other arguments until we get Glenn to admit his errors… (though I doubt we ever will)

  • Glenn March 6, 2012, 6:29 pm

    Stephen, Glenn “finally realizes” that he needs to cut his premises down from five in order to escape trouble?

    Stephen, you can’t mean that. This claim on my part that I am using three premises is not a back-track. Cast your mind back to when I first alerted you to that blog post on the simple explanation of the moral argument. I pointed out to you then that while as written there were five steps, in fact it is reducible to two premises and a conclusion. In fact, what you’ve quoted is me reminding you that it only required two premises – reminding you that I had stated this some time ago. And yet you introduced this quote as though this was a new manoeuvre on my part to get out of trouble. To now claim that this is a new move, a sort of back-track, is not really very honest.

    And OK, now you’ve admitted that your claim about probability was flat wrong. Nice to see! You wouldn’t do that before, instead trying to minimise or re-portray the error, saying that really it was just not “tightly” worded. Now it has been upgraded to flat wrong, which is progress. Your claim about probability was dead wrong, but once this was shown to you, you still seemed to be buoyed by your initial hope that you’d found a great rebuttal. Even now that the probabilistic argument has been discarded, it still appears to be playing a crucial role, and has really become your major reason for not thinking that the moral argument is any good.

    And now, with all respect, it has become clear that you’re attacking straw when you complain about me failing to acknowledge that you’re right about a hypothetical five premise argument. Here’s why: I never claimed that your hypothetical five premise argument was strong. Did I? If so, where? So why now – when I raise a concern about partisanship, would you now suddenly engage in a tu quoque by saying “Oh yeah? Well you never admitted that my own hypothetical argument isn’t very strong!”

    So what? A hypothetical like that isn’t impressive either way. Your hypothetical argument would establish that the actual probability of a claim is between 32% and 100%, which is neither awful nor awe-inspiring. But I never said that was a very strong argument, so why you think I now have to admit that it’s not very strong… I guess you’ve lost me. Matt is on the money here: “Glenn, please acknowledge that an argument you NEVER made was weak, and that Stephen is right about a trivial claim.” But for what it’s worth: An argument like that does not establish an impressive lower bound of probability, no. It just shows that a conclusion isn’t very unlikely. If for any reason you think that I have denied that somewhere, then you now know that I do not deny this (and to my knowledge, I never have).

    Is that clear enough? What relevance this has does escape me somewhat…

    “Fail to acknowledge that I was RIGHT when I said I had made this point prior to your saying I hadn’t”

    No. I did not fail to acknowledge that you had made this point about your own hypothetical argument. Not sure what you’re getting at here, but I acknowledged that you had made this point. That’s why I kept adding that this wasn’t important, since my argument isn’t a five premise argument (remember how I kept saying that in order to be relevant when it comes to my argument, 0.32 would need to change to 0.64). So while I’m happy to admit that your hypothetical argument wasn’t impressive, I have no reason to admit that I had denied that you had made this point before. You made it repeatedly, apparently gaining confidence from doing so.

    And I’m surprised to see you claiming that I had failed to acknowledge that the conditional premise requires extra support. Goodness, you complained about how long my blog post was when I tried to lay out just that support. How can I “acknowledge” this without simply ignoring the facts? Now – and here is the rub, Stephen, so don’t miss this – you might personally believe that the “rambling” case I made for that premise doesn’t warrant me treating it as though it has a probability of 0.8 (or any high probability). Or you might reject every single argument that doesn’t include every reason for accepting every premise (I think this is unreasonable). But as it stands, I have always acknowledged that the premise needs support. Obviously the level of certainty that I have about that premise is meant to be the certainty that one can – in my view – have in light of the reasons I have given for believing it. So let’s have none of this “Glenn won’t admit that the premise needs supporting premises” stuff.

    And as for the last thing: “Fail to acknowledge that the “upper bounds” point was almost entirely smokescreen”

    I never said anything about upper bounds. My “lower bound” point was anything but a smokescreen. It was an explanation of why your probability claim was “flat wrong,” something we now agree on.

    This is now becoming a drawn out case of you getting so tied up in trying to find a “naughty list” that the big issues are slipping away. Unfortunately, I wonder if the real aim here is to distract from the issues that have been spelled out. Sorry if that’s unpleasant, but that’s how it looks. So to recap:

    1) The only saving grace you seemed to think you have for the EGC in all of this was that 1) we’re not allowed to consider the moral argument until the problem of evil is fully dealt with first (an ad hoc requirement if ever there was one)
    2) The only actual reason you could conjure up for rejecting theodicies is that they are ad hoc (a claim that I believe I have shown to be obviously false in a more recent blog)
    and
    3) The only actual substantial criticism that I have seen you give of the moral argument itself was based firstly on a method of probability assessment that you now find to be “flatly wrong,” and more recently on a method of assessment that involves simply insisting that I include more premises, a move that appears to just assume without argument that whatever reasons I have laid out in support of the conditional premise of the moral argument fail to give it the credibility that I treat it with. Sorry, you’ll need to argue for this.

    So your protracted distractions nothwithstanding, the upshot of all this is that the Evil God Challenge has not fared very well. When you add to this the other considerations that I have offered against it (not counting the moral argument), I’d say the most respectable thing to do would be to shelve it.

  • Dave March 6, 2012, 8:04 pm

    Ow, that’s going to leave a mark! I think that’s a helpful summary Glenn. I really am getting the feeling that the more Stephen says lately, the more it looks like he’s getting people to look away from the meat of the arguments.

  • matt March 6, 2012, 10:56 pm

    Stephen,

    “Sure I just said above that Glenn sets out a case for the premises, didn’t I?”

    and then

    “Moreover, even when Glenn tried to make a case for the conditional premise, Glenn never set out the actual premises so we could see what they were. I asked him to. He refused. We all know why….”

    What this sounds like to me is that you are saying, “I asked Glenn for six and he robbed me! He only gave me half a dozen!”

    but whatever, it’s a done deal.

  • Nathan March 7, 2012, 12:04 am

    Matt – I am not dealing with any other arguments until we get Glenn to admit his errors… (though I doubt we ever will)

    Sounds like an escape plan! Shame really…

    Matt’s comment (#40) is so insightful, I’d like to nominate it as the basis for the next Praxman script!

  • Calum Miller March 11, 2012, 7:56 am

    One would be justified in assigning a probability higher than 0.32, provided it were at least logically possible that one of the premises be false and yet the argument still true. We can’t say how much more probable, but *that* we can say it is more probable follows directly from the probability calculus.

  • Mike D March 13, 2012, 5:15 pm

    Bringing up the moral argument in an argument about whether God is good or evil strikes me as a red herring. The moral argument is essentially a question of meta-ethics – what our basis is for defining terms such as “good” and “evil”. But the Evil God Argument only requires that we accept conceptualizations of good an evil as generally understood by Christians. That’s enough to demonstrate that 1) God’s goodness cannot be demonstrated any more (or less) than God’s evilness, rendering claims about God’s nature indeterminable and thus absurd, and 2) as a corollary, that the Christian presumption of God’s goodness is without logical foundation.

  • Glenn March 13, 2012, 7:21 pm

    Mike D, it’s not a red herring for the following reason: If it’s true that God is the basis of all moral duties, then God’s nature or will is revealed in those duties. If those duties involve things that human beings think of as good (e.g. kindness, mercy etc), then God’s nature or will is good in the way that human beings think of goodness. If that’s true, then God is not evil but good.

  • matt March 14, 2012, 9:00 am

    Mike, I want to differ with your statement that the EGC merely requires that we accept traditional theistic conceptions of good and evil. It most certainly requires that we avoid a traditional theistic conception of evil in order to go all the way through, since, on the traditional conception of evil, God most certainly cannot be anything but good. He must be Goodness all the way, and that is gotten around to by metaphysical considerations about the world, the degrees to which things have various properties like being good, powerful, loving, etc. God, if he is God, would be where the buck stops, so to speak.

  • Giles March 9, 2014, 8:25 pm

    Well, on the Divine command theory I think the “Evil God” is incoherent as God gets to define good and evil. But then on that theory, as C.S.Lewis pointed out, all one is saying when one says God is good is that he wills what he wills. As does Satan. So it all comes down to God being omnipotent and Satan not. Or put crudely “might is right!” The essence of Calvinism, no need for a theodicy when God can play Humpty Dumpty.

  • Glenn March 9, 2014, 8:31 pm

    Giles, the Divine Command Theory does not mean that God defines good by an act of will. It means that right and wrong – that is, moral obligation and prohibition, is what God commands and forbids. That’s quite different, hence C.S. Lewis’ criticism, as you have summarised it, does not succeed.

Leave a Comment

Remember: All comments should conform to the blog policy and you must use your real name. Comments that do not conform may be removed in whole or in part. You can review the blog policy here.

Current ye@r *

 Characters remaining