Rescuing Dawkins’ Case Against God (part 2)

by Luke Muehlhauser on January 6, 2010 in Criticism of Atheists

See the index.

richard dawkins

Last time, I wrote that Richard Dawkins was philosophically naive to assert that:

To explain [something] by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer.1

You cannot require that explanations themselves be explained, for this leads to an infinite regress of explanations, and then we would never be able to explain anything.

Such a requirement would literally destroy the natural sciences that Dawkins loves so much. Indeed, most of our scientific explanations open up new puzzles and questions. This is recognized by philosophers of science and famous atheist philosophers.2

I also quoted William Lane Craig’s analogies about arrowheads found by archaeologists and machines found on the back of the moon. In order for “Intelligent beings designed them” to be the best explanation for arrowheads in the dirt and machines on the moon, we do not need to also explain where the intelligent beings came from.

Objections

Many of my readers suffered from a simple misunderstanding, here. They thought that Craig’s analogies were meant to show that God could be the best explanation for our complex world. But that wasn’t the point at all. Craig wasn’t arguing for God’s existence, there. His point was merely that we can’t require that good explanations themselves have be fully explained, which is true.

Some readers simply denied my assertion. They said that all good explanations are themselves, explained. Psychological explanations are explained in terms of biology, biological explanations are explained in terms of chemistry, chemical explanations are explained in terms of physics, etc.

While this may be the ideal, this is not how science works. Good chemical explanations were offered for things before we knew anything about “atoms.” Today, scientists posit non-baryonic matter (“dark matter“) to explain certain observed features of galaxy rotation and clustering, even though they have no explanation whatsoever for the dark matter itself. And psychologists regularly explain behavior in terms of theoretical structures for which the underlying neuroscience is not the least bit understood.

So it’s simply false that successful explanations must themselves be explained. To make such a requirement would lead to an infinite regress of explanations, and we would never have an explanation for anything.

Other readers, like Answers in Genesis Busted, wrote that my response to Dawkins is inadequate because Dawkins’ argument can be reinterpreted so that it doesn’t depend on the flawed principle that every good explanation must itself be explained.

But of course! I only intended to show that one of Dawkins’ major points is false. So let us now explore the merits of his argument more thoroughly.

Lipton, Inference to Best Explanation, 24. The same argument, as it hap-
pens, is put into the mouth of Cleanthes in David Hume’s “Dialogues” (iv
[65]; see Mackie, Miracle of Theism, 143).


Dawkins’ argument

On pages 156-157 of The God Delusion, Dawkins summarizes his main argument for atheism:

This chapter has contained the central argument of my book, and so, at the risk of sounding repetitive, I shall summarize it as a series of six numbered points.

1. One of the greatest challenges to the human intellect… has been to explain how the complex, improbable appearance of design in the universe arises.

2. The natural temptation is to attribute the appearance of design to actual design itself…

3. The temptation is a false one, because the designer hypothesis immediately raises the larger problem of who designed the designer… It is obviously no solution to postulate something even more improbable [than the design itself]…

4. The most ingenious and powerful [explanation of complexity] so far discovered is Darwinian evolution by natural selection…

5. We don’t yet have an equivalent [theory] for physics. Some kind of multiverse theory could in principle do for physics the same explanatory work as Darwinism does for biology…

6. We should not give up hope of a better [theory] arising in physics, something as powerful as Darwinism is for biology…

[Therefore] God almost certainly does not exist.

The obvious problem here is that Dawkins’ argument is totally invalid. No laws of logic would allow you to deduce that conclusion for those premises.

Still, perhaps we can radically reformulate Dawkins’ reasoning into an argument with some force. We will attempt this in the next post.

  1. The Blind Watchmaker, page 141. []
  2. For example, see Peter Lipton, Inference to Best Explanation, page 24; J.L. Mackie, Miracle of Theism, page 143. []

Previous post:

Next post:

{ 120 comments… read them below or add one }

Torgo January 6, 2010 at 6:02 pm

I see how having an infinite regress of explanations would leave things unexplained, but doesn’t science continually strive to dig deeper, to find explanations for the explanations, and so on? I understand that in the short-term, science uses place-holder explanations like dark matter, but it doesn’t deliberately stop at them, it tries to better understand them. So, yes, science need not have an explanation of every explanation in order to keep going. But as a whole, isn’t science moving in the direction of an infinite regress of explanations? Unless it hits upon something that can’t be further explained or investigated scientifically–a point which we seem to be approaching.

  (Quote)

Sharkey January 6, 2010 at 6:09 pm

Luke, thanks for responding to my comment.

I guess I’m still confused by the phrase, a “good explanation”. What would be considered a good explanation of a scientific problem? Is a good explanation a) true, b) mostly correct, or c) just plausible?

In the first two cases, I think my point stands. In the last case, I’m unsure.

  (Quote)

Moe January 6, 2010 at 6:17 pm

Sharkey has already touched on what I was going to ask. Why should we grant that Craig’s explanation is GOOD at all? What defines a GOOD explanation?

  (Quote)

Andrew Brenner January 6, 2010 at 6:24 pm

When I first encountered Dawkin’s argument I thought it was simply awful. However, to be fair, it seems that as prestigious a philosopher as J.J.C. Smart has offered a somewhat similar argument:

“If we postulate God in addition to the created universe we increase the complexity of our hypothesis. We have all the complexity of the universe itself, and we have in addition the at least equal complexity of God. (The designer of an artifact must be at least as complex as the designed artifact) … . If the theist can show the atheist that postulating God actually reduces the complexity of one’s total world view, then the atheist should be a theist.” – “Laws of Nature and Cosmic Coincidence,” The Philosophical Quarterly 35 (1981), 275-276

Now, of course, this argument is still awful. That the conjunction of explanans and explanandum in any successful explanation should be simpler than the explanandum itself is a bizarre suggestion. How exactly would that work? And wouldn’t most explanations we are otherwise inclined to accept be hopelessly complex? (You will, of course, be forbidden from believing that the explanation for these typed words is me, Andrew, another human being, because then you’d have all the complexity of the words themselves, and on top of that all the complexity of a human being. Yikees!)

Also, recall that Smart writes: “If the theist can show the atheist that postulating God actually reduces the complexity of one’s total world view, then the atheist should be a theist.” Doesn’t Berkeley’s theistic idealism satisfy this requirement? And isn’t that, well, more or less a reductio of Smart’s position?

  (Quote)

Cafeeine January 6, 2010 at 7:23 pm

I think the problem that has so many people riled up is that the explanations offered in the examples are not analogous to the god question.

For one thing, the explanations to the arrowheads and the machinery are tentative and have the potential of getting further explained, and possibly being proven false with further evidence.
Postulating a supernatural entity beyond verification as the final explanation will necessarily shut down further inquiry. As such, the standard for accepting it would be considerably higher

There is a difference in accepting a tentative explanation without itself having a good explanation, and doing the same for a final explanation, in my opinion.

  (Quote)

Ben January 6, 2010 at 7:31 pm

In order for “Intelligent beings designed them” to be the best explanation for arrowheads in the dirt and machines on the moon, we do not need to also explain where the intelligent beings came from

No we don’t, but we do need to explain why we decided that intelligence was necessary in the first place.This is the real sticking point for religions, because one of their main arguments is that complexity requires intelligence, the universe is complex, therefore intelligence created it. But surely, if “God” can create the universe in no time at all, he is more complex than the universe. If complexity requires intelligent design, and “God” is complex, then “God” required a designer.

Sure, this doesn’t dispute your main argument that infinite regression does not lead to the conclusion that some form of deity is “probably” not possible, but this is only at the most basic philosophical argument for deism. The vast majority of religious believers will clearly state that their god is the end of the line, that it created everything but did not need to be created itself.

I think it is these deists that Dawkins is targeting his argument towards, simply because they are the ones using complexity as an argument for design while also claiming that design ends with “God”.

It’s still possible to use design of complexity as an argument, but not while disallowing infinite regression as an inherent attribute.

  (Quote)

lukeprog January 6, 2010 at 7:38 pm

Sharkey,

Explanation is a huge topic in philosophy. Tom Gilson and I talked about it a fair bit.

  (Quote)

Antiplastic January 6, 2010 at 8:47 pm

Andrew Brenner: That the conjunction of explanans and explanandum in any successful explanation should be simpler than the explanandum itself is a bizarre suggestion. How exactly would that work? And wouldn’t most explanations we are otherwise inclined to accept be hopelessly complex?

I can’t imagine a question more utterly misconstruing how empirical explanations work.

When you think about it, all good empirical explanations with equivalent predictive powers have the property of being less complex than the initial object to be explained. To explain is just to say that it’s easier to model [Hypothetical initial conditions + Bridge Laws] than it is to model [Explanandum]. If your explanation is working right, the description of the observation squirts out the end whenever you run the “Hypothetical initial conditions + Bridge Laws” calculation, so you don’t need to waste processing power storing it in your total description.

Of course, what we get from even the strongest conclusion of first cause arguments is…, well, what exactly? An indescribably complex initial condition, a nomological relation that amounts to simply reiterating “initial condition is, like, totally capable of making this observation!”, and then basically nothing. Its predictive power is zilch. The reason even theistic physicists who believe the KCA don’t ever ever actually use it in their work is because it entails no predictions other than the one single observation it was jury-rigged to squirt out.

(eta yes I am aware of the irony of my using “calculation” loosely when only the other day I was railing against that practice. But hopefully this helps more than it hurts.)

  (Quote)

Hansen January 6, 2010 at 8:53 pm

The Creator Hypothesis is an infinite regression not because the creator is unexplained but because it doesn’t explain anything. It’s an arbitrary “placeholder” explanation. Is it actually any better than saying that we are all part of a computer simulation that started five minutes ago, provided with ready-made memories, with holes in our socks and hair that needs cutting?

  (Quote)

Paul January 6, 2010 at 8:58 pm

Cafeeine: I think the problem that has so many people riled up is that the explanations offered in the examples are not analogous to the god question.

The 2nd part of this comment is what I was trying to articulate in my first comment to Luke’s original post. I think you succeed better than I with a lot less words

  (Quote)

Conversational Atheist January 6, 2010 at 8:59 pm

I quickly read Dawkins’ book a number of years ago, and I guess I was being charitable in my memory on this argument.

If a person who advocates ID via some argument like: “something more complex cannot come from something less complex — and the more complex something is, the less likely it is to exist spontaneously; therefore it is likely that God created this complex universe/humans/whatever” …then, a Dawkins-like argument would force the person to accept one of two horns in a dilemma: A) Give up the “more complex cannot come from something less complex” premise or B) Give up on God being a more likely explanation for the universe/humans/etc.

  (Quote)

Paul January 6, 2010 at 9:07 pm

“The obvious problem here is that Dawkins’ argument is totally invalid. No laws of logic would allow you to deduce that conclusion for those premises.”

I agree that the conclusion doesn’t follow from the premises.

Perhaps Dawkins should use a variation of Richard Carrier’s analogy of a horse that has won every race he has been in.

For me, whether Dawkins articulates things wrong or deduces conclusions improperly my take away is that saying “God did it” is not an explanation at all. It is an assertion with no supporting evidence. And perhaps when you get down to it that might be what Dawkins ultimately means.

  (Quote)

lukeprog January 6, 2010 at 9:10 pm

Cafeeine and Paul, did you read this paragraph?

Many of my readers suffered from a simple misunderstanding, here. They thought that Craig’s analogies were meant to show that God could be the best explanation for our complex world. But that wasn’t the point at all. Craig wasn’t arguing for God’s existence, there. His point was merely that we can’t require that good explanations themselves have be fully explained, which is true.

  (Quote)

Paul January 6, 2010 at 9:15 pm

lukeprog: Cafeeine and Paul, did you read this paragraph?
  

Fair enough. Seems I read into it more that was intended. I’ve agreed from the get-go that the best explanation itself doesn’t require an explanation. I think perhaps where I over extended myself was in thinking that Craig was implying that his analogy was apples to apples comparison.

  (Quote)

Cafeeine January 6, 2010 at 9:33 pm

@luke.
I did read that. I was simply trying to clarify the point of contention, and possibly point out that there is a difference between tentative explanations and final explanations. IN effect, I believe that WLC tried to shift the focus of RD’s statement, to avoid addressing this difference.
When RD said this:
“To explain [something] by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer”

The focus is on the ‘supernatural’ part, that in the Abrahamic creation stories is a final answer, the end to inquiry, not at all like the tentative conclusion of the archeologists explaining arrowheads.

I will admit to being a rank amateur in matters of philosophy, so there may be some important issue of form thst escapes me, or that also escaped RD, but it seems to me that WLC purposely threw that argument as a red herring.

  (Quote)

Andrew Brenner January 6, 2010 at 11:32 pm

Antiplastic:
When you think about it, all good empirical explanations with equivalent predictive powers have the property of being less complex than the initial object to be explained. To explain is just to say that it’s easier to model [Hypothetical initial conditions + Bridge Laws] than it is to model [Explanandum]. If your explanation is working right, the description of the observation squirts out the end whenever you run the “Hypothetical initial conditions + Bridge Laws” calculation, so you don’t need to waste processing power storing it in your total description.  

ehhh? That doesn’t seem true at all (not to mention it isn’t exactly relevant to the idea I took Smart to be advocating). In any case I can think of plenty of good empirical explanans that are more complex than the explanandum they are supposed to explain. Whether the explanans is a good one is, i would say, a factor of its ability to predict the explanandum (i.e. the probability of the explanandum given the explanans) and its prior probability. The prior probability of the explanans will no doubt be determined in part by its simplicity, but *only* in part. Once we concede that obvious point there’s simply no justification for saying a good explanans will *always* be simpler than the explanandum — there are other factors involved.

Example: I find a letter in my mailbox. I explain the letter by invoking highly complex human beings who wrote and delivered the letter. This is a good explanation in part because it has a high prior probability based on the fact that it meshes well with my background experience (every past letter in my mailbox was written and put there by a human being, let’s say). The complexity of the explanans (that is, the complexity of the physical/physiological makeup of the posited human beings – *not* complexity in terms of failing to mesh well with our background knowledge) is a non-issue.

  (Quote)

mikespeir January 7, 2010 at 3:36 am

Nonexistent entities don’t explain anything. First show that the entity exists and then we can talk about whether it will serve as an explanation.

  (Quote)

Reginald Selkirk January 7, 2010 at 5:49 am

Craig wasn’t arguing for God’s existence, there.

Quite true. Note that he never proposed, as the origin of his putative pottery shards, or fictional moon machinery, “God did it.” Since he is a theist, the omission is curious. He certainly believes in the sort of omnipotent, interactive God Who could do such a thing. Yet even he recognizes the ridiculousness of such a “solution.”

  (Quote)

Reginald Selkirk January 7, 2010 at 6:03 am

Dawkins should have spent more time consulting with someone trained in logic before offering up such a train wreck of an argument as the centerpiece of his book.

And Craig should have spent time consulting someone trained in actual archaelogy before offering such a tran wreck- strawman version of it in his counter-argument. And yet you completely overlook that. Shame on you.

  (Quote)

Reginald Selkirk January 7, 2010 at 6:09 am

I also quoted William Lane Craig’s analogies about arrowheads found by archaeologists and machines found on the back of the moon. In order for “Intelligent beings designed them” to be the best explanation for arrowheads in the dirt and machines on the moon, we do not need to also explain where the intelligent beings came from.

You most certainly do. I could propose that the arrowheads were constructed by Na’Vi. certainly it would weigh against my putative explanation that a) Na’Vi are completely fictional b) That even if they existed, they have never been to planet Earth.

Your naivety of archaeology matches Craig’s.

  (Quote)

Hansen January 7, 2010 at 6:50 am

Considering that you are posting only a summary of Dawkins’ argument for the improbability of the existence of God, you ought to at least quote his own summary completely. It’s not actually that long (emphasis mine):

3. The temptation is a false one, because the designer hypothesis immediately raises the larger problem of who designed the designer. The whole problem we started out with was the problem of explaining statistical improbability. It is obviously no solution to postulate something even more improbable. We need a ‘crane’, not a ‘skyhook’, for only a crane can do the business of working up gradually and plausibly from simplicity to otherwise improbable complexity.

Both you and Craig seem to be very focused on the phrase “who designed the designer”. I guess this is somewhat forgivable since Dawkins himself has occasionally used this phrase as a “soundbite” in various interviews. He usually also calls it a “nonstarter” and explains that it is of no use postulating something even more complex in order to explain complexity.

Personally, I think Dawkins is perhaps focusing a bit too much on saying that the designer raises a larger problem or that it is even more improbable. Although I agree that the designer hypothesis is worse than no hypothesis at all, the real issue for me is that it completely lacks explanatory power; it is truly a nonstarter.

  (Quote)

Reginald Selkirk January 7, 2010 at 6:53 am

It is ironic that Dawkins is being accused of philosophical naivety, but in making the charge, Craig – and by extension you, for repeating him uncritically – are making scientific arguments with scientific naivety.

  (Quote)

Dan Gilbert January 7, 2010 at 8:09 am

I think the problem I have with Craig’s “the explanation doesn’t need an explanation” claim is one of plausibility. If the explanation proposed is implausible or completely without any kind of rational basis, then it DOES need an explanation.

Example: My drive into work today took longer than usual because of excessive cloudiness in a neighboring county.

I certainly think that my “explanation” for my long drive needs some explaining because it’s implausible. It makes no sense without some further explanation (after which, it may still make no sense). It would be perfectly reasonable to dismiss it… unless a rational explanation was given for it.

If Craig’s statement is valid, then anything can be submitted, unchallenged, as a valid explanation.

I do agree, however, that Dawkins’ “logic” path shown in this article is… ummm… not logical (though I agree with his final statement on its own). ;-)

  (Quote)

lukeprog January 7, 2010 at 8:13 am

Reginald,

I think Craig’s illustrations for why you needn’t have an explanation of an explanation are perfectly useful. I gave further illustrations of my own. Seriously, this is a really basic principle of philosophy of science.

  (Quote)

lukeprog January 7, 2010 at 8:15 am

Reginald,

The reason the Na’vi are a poor explanation for arrowheads found in the dirt is NOT because we have no explanation of the Na’vi, but because such an explanation fails on several other explanatory virtues. Your naivety of philosophy of science matches Dawkins’.

  (Quote)

lukeprog January 7, 2010 at 8:18 am

Hansen,

You are right that I have so far focused on “Who Designed the Designer.” That is why I am continuing this series by examining Dawkins’ argument from more angles.

  (Quote)

Hansen January 7, 2010 at 8:36 am

lukeprog: Hansen,You are right that I have so far focused on “Who Designed the Designer.” That is why I am continuing this series by examining Dawkins’ argument from more angles.  

That’s fine with me. Although I think that Craig’s infinite regress counter-argument is invalid, it is important to show why it is invalid.

  (Quote)

drj January 7, 2010 at 8:52 am

Luke, again I think you are giving short rift to the role of complexity in Dawkins’ argument. Of course, this could just be because Dawkins’ was sloppy. However, unlike Craig and yourself, Alvin Plantinga has made several responses to Dawkins’ arguments, and he seems to understand the role of complexity in a similar way that I do. So I don’t think its just me here.

It seems to be a given, from both theists and atheists alike that seemingly organized complexity requires explanation. Evolution is such an explanation, and ID is such an explanation. Dawkins contends that God must be an even more complex form than the universe or any of the complexity in it – hence, he requires an explanation.

  (Quote)

drj January 7, 2010 at 9:01 am

And Dawkins certainly doesnt tackle head on any arguments about divine simplicity – perhaps that would more appropriately be called a naive omission, on his part.

  (Quote)

drj January 7, 2010 at 9:27 am

I also agree with Reginald in the arrowhead example. Concluding design most definitely requires a set of explanatory background information. Absent that background information (or even in the case of not enough background information), you simply have an assumption, or an axiom, but nothing that yet rises to the level of explanation.

I suppose a regress of explanations will be required until things are reduced to an axiom or some sort of deductive explanation

  (Quote)

MauricXe January 7, 2010 at 10:58 am

I agree with Ben and Hansen.

Dawkins’ argument is not a simple quip like “NOU!” His argument, taken in its entirety, is that the presence of complexity does not entail a designer. If it does, then a complex designer has to have a creator, or at least one can pose the question.

Craig is simply wrong here to assert that Dawkins merely claims a designer needs a designer. Taken out of context, sure, it does seem that is what Dawkins is claiming BUT it isn’t. His argument, in this regard, is a strawman.

Craig does explain, in some of his other work and debates, that God is a immaterial mind. Why not just stop at “God did it” instead of explaining it? It may very well be the case that God does exist and that he created the universe just as Craig believes. But one does need an explanation of his existence to even consider this possibility. Ofc there will be a point when one has to say “it just is that way” but that doesn’t mean we can’t investigate the intermediate steps.

  (Quote)

Silver Bullet January 7, 2010 at 11:36 am

I think I disagree with you Luke.

Let’s look at what Dawkins says:

1.”One of the greatest challenges to the human intellect… has been to explain how the complex, improbable appearance of design in the universe arises.”

2.”The temptation is a false one, because the designer hypothesis immediately raises the larger problem of who designed the designer… It is obviously no solution to postulate something even more improbable [than the design itself]”

While Dawkins hasn’t phrased his argument in a deductive format, I do think that he is making a good argument. He is in fact arguing that a particular ID argument is self-refuting.

The ID argument that Dawkins is addressing goes like this:

As he articulates in point # 1 above: the universe and life in it is too complex to have arisen out of chance, so it must have had an intelligent designer.

However, as Dawkins argues in point # 2 above, a creator god must be even more complex, and even less probable than the universe whose complexity and improbability the creator god is meant to explain. Accordingly, such an argument is self refuting – it can’t provide any explanation of what it is proposed to explain.

There may be other arguments for ID that Dawkins does not address, but the one he does, which addresses the issues of improbability and complexity (the appearance of design), does, I think, work well. Accordingly, I also think that Craig’s is a strawman based on only a portion of Dawkin’s argument.

  (Quote)

Paul January 7, 2010 at 12:04 pm

I think the issue many of us are having or have had is that we are reading a motive into Craig’s analogies. I don’t know if what we were reading (or at least I was) into Craig’s motive is truly valid or not. If we set this aspect aside for sake of discussion I think most if not all who have commented agree with the premise that a best explanation itself does not need an explanation. What it needs is the reasoning for why it is the best explanation.

I believe Luke has made it clear that he doesn’t think that God is a best explanation.

Where I think I may have a slight disagreement with Luke is in one interpretation.

“…because the designer hypothesis immediately raises the larger problem of who designed the designer… It is obviously no solution to postulate something even more improbable [than the design itself]…”

Whereas, at least in these series of posts, I think Luke has a strict reading of this quote, I take it more metaphorically. But then maybe those are my biases seeping in.

  (Quote)

lukeprog January 7, 2010 at 2:33 pm

drj,

Yeah, I haven’t even gotten to speaking about complexity yet. That’s coming later.

  (Quote)

Shane January 7, 2010 at 3:31 pm

Luke, I am afraid you have slipped up here; you have allowed the word “explanation” to shift track mid-argument. Dawkins was writing a popular book for the mass market, and did that very very well. I accept that some philosophers may smart at his to-the-point text, but that’s too bad.

Humans are a *given* in the explanatory system for arrowheads or watches on heaths. We can see humans and interact with them. Yes, they require explanations, but they remain a given. So in order to explain the arrowhead we do NOT need to dig any further; we have done the job with respect to the arrowhead; further digging is in order to explain a DIFFERENT problem. That’s the switch.

As far as the universe is concerned, the situation is quite different. A designer is NOT a given; it is an ad hoc explanation, the postulation of which raises far more questions than it answers.

So I suggest that Dawkins has this right (and I fully support the language he used to express it), and you and Craig have this wrong. A bit more on my blog at http://answersingenes.blogspot.com

I still love your blog btw :-)

  (Quote)

lukeprog January 7, 2010 at 5:56 pm

Shane,

I don’t know what else to say except to repeat for about the 12th time that neither Craig nor I were, in those analogies, defending the idea that God or a designer is a good explanation for arrowheads. That was not the point. The point was solely to show that in order for something to be the best explanation of something, one need not also have an explanation of the explanation.

  (Quote)

Cafeeine January 7, 2010 at 6:28 pm

Luke,
In my opinion that point made by both you and Craig is sound on its own, however it is misplaced as a response to Dawkins. Craig is using it as a red herring.

Dawkins didn’t say “To explain [something] by invoking X is to explain precisely nothing, for it leaves X unexplained” for every value of X, in which case Craig’s point would have been a valid response, but he specified a “supernatural designer” that is a particular explanation with particular properties. In short,

Craig argued against a point RD wasn’t contesting, in order IMO to somewhat equate the ‘tribe’ conclusion for the arrowheads, to the divine conclusion for the universe. It was a completely unnecessary statement.

  (Quote)

lukeprog January 7, 2010 at 6:39 pm

Cafeeine,

Dawkins said that “God did it” is a poor explanation for complexity BECAUSE it leaves God unexplained. Now it is true that “God did it” is a poor explanation for complexity, but this is NOT because it leaves the explanation itself (God) unexplained.

  (Quote)

W January 7, 2010 at 6:41 pm

This is just more of Luke giving Craig too much credit again. Everyone is focused on the bestexplaination part and not focusing on the important word here, SUPERNATURAL explaination. Dawkins IS RIGHT to say a supernatural explaination explains nothing, anyone who disagrees can kindly provide me an example where supernatural causes are useful explainations for a given phenomena. I guess we should have just taken demons as the cause of disease, since it was the best explaination.

  (Quote)

Cafeeine January 7, 2010 at 7:08 pm

Luke,

I think it is a factor specifically because the proposed explanation is a god, i.e a supernatural explanation that closes off inquiry.

However I believe you said you’ll address this issue more, I’ll wait and see what you have to say and if need be comment again. It may just be that we are approaching the issue from different angles, or that I’m hopelessly wrong.

  (Quote)

Tony Hoffman January 7, 2010 at 7:34 pm

Luke: The point was solely to show that in order for something to be the best explanation of something, one need not also have an explanation of the explanation.

You are also making it sound as if this is something that Dawkins is confused about, and you appear to be lending credence to Craig’s diversionary attack on Dawkins’ (valid) criticisms of theistic explanations.

There was an ad during the ’80′s that had Kelly LeBrock in it, and it began with Kelly saying “Don’t hate me because I’m beautiful.” Lynn Barry (the comic strip artist) was on Letterman, and she mentioned the commercial, and said simply, “That’s not why I hate you.”

Dawkins’ doesn’t hate the theistic explanation of the why is there something instead of nothing because it leaves God unexplained. He hates it because it doesn’t answer the question.

  (Quote)

lukeprog January 7, 2010 at 8:35 pm

W,

I agree that Dawkins is right to say that “God did it” is a bad explanation. But Dawkins said it’s a bad explanation BECAUSE it leaves the explanation (God) unexplained. THAT is the part that is false, and the part I argue against. Dawkins needs to give a TRUE account of why “God did it” is a poor explanation, not a false and misleading one. If Dawkins’ reason for rejecting the God hypothesis is because it leaves the explanation unexplained, then Dawkins has given no reason at all.

  (Quote)

lukeprog January 7, 2010 at 8:37 pm

Tony,

No, you’re wrong. Dawkins explicitly says that God is a bad explanation because it leaves the explanation itself unexplained. Read Dawkins’ quote again. That is EXACTLY what he says.

  (Quote)

Silver Bullet January 7, 2010 at 9:09 pm

luke: Dawkins explicitly says that God is a bad explanation because it leaves the explanation itself unexplained. Read Dawkins’ quote again. That is EXACTLY what he says.  

It seems to me that as an explanation for the origin of everything, the god argument does fail exactly because it simply begs the question of the origin of god.

Thus, while Craig may be correct when suggesting that an explanation should not be discarded because the argument itself requires an explanation, the explanation for everything surely must.

  (Quote)

W January 7, 2010 at 9:23 pm

And reading through the responses do nothing but make me respect you less, Luke. To denegrate people for a lack of knowledge only shows YOUR ignorance. I know Craig’s really the only competition left for atheists, so you might be a little reluctant in seeing him fall, but to help keep him afloat by ignoring HIS errors and emphasizing other’s is wrong. Dawkins point was that a SUPERNATURAL explaination is a non-explaination. Not only is this a straw-man rebuttal, but like others have said, a red herring. Luke, you are not helping atheism at all with this anti-hero worship.It’s sad to see, considering how clear and correct you are on other subjects, but when Craig comes up, seems the rosy colored glasses get put on.

  (Quote)

drj January 7, 2010 at 9:34 pm

lukeprog: I agree that Dawkins is right to say that “God did it” is a bad explanation. But Dawkins said it’s a bad explanation BECAUSE it leaves the explanation (God) unexplained. THAT is the part that is false, and the part I argue against. Dawkins needs to give a TRUE account of why “God did it” is a poor explanation, not a false and misleading one. If Dawkins’ reason for rejecting the God hypothesis is because it leaves the explanation unexplained, then Dawkins has given no reason at all.

I think Dawkins would say God is a bad explanation because God must be complex. Not just any kind of complexity – specified complexity. If one is borrowing the logic of design arguments, than this specified complexity must itself be explained by a an even more complex designer. Hence, the infinite regress of designers.

  (Quote)

TaiChi January 7, 2010 at 10:04 pm

Like others here, I think Luke is misconstruing Dawkin’s argument. Here is how I read him:

“3. The temptation [to postulate a designer] is a false one, because the designer hypothesis immediately raises the larger problem of who designed the designer. [Which is problematic because..] The whole problem we started out with was the problem of explaining statistical improbability. It is obviously no solution to postulate something even more improbable. [So, contrary to this inclination..] We need a ‘crane’, not a ‘skyhook’, for only a crane can do the business of working up gradually and plausibly from simplicity to otherwise improbable complexity.”

I think that’s the obvious and natural way to read this point, whereas on Luke’s interpretation, the “who designed the designer?” tu quoque is point 3a and and the “a designer would be even more improbable” would be point 3b, erroneously lumped together under point ’3′.

Well, maybe they could be different points. Since Luke quotes The Blind Watchmaker, then I’d hope that in this other context that would be clear and justify his view. But the passage reads…

“Once we are allowed simply to postulate organized complexity, if only the organized complexity of the DNA/ protein replicating engine, it is relatively easy to invoke it as a generator of yet more organized complexity. That, indeed, is what most of this book is about. But of course any God capable of intelligently designing something as complex as the DNA/protein replicating machine must have been at least as complex and organized as that machine itself. Far more so if we suppose him additionally capable of such advanced functions as listening to prayers and forgiving sins. To explain the origin of the DNA/protein machine by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer. You have to say something like ‘God was always there’, and if you allow yourself that kind of lazy way out, you might as well just say ‘DNA was always there’, or ‘Life was always there’, and be done with it.”

.. and reverses the order of the very same supposedly separate points Dawkins makes in the God Delusion. That’s a pretty fair indication that Dawkins thinks 3a and 3b go together.

  (Quote)

TaiChi January 7, 2010 at 10:25 pm

“Even still, it’s hard to see how he could ever conclude that “God almost certainly does not exist” from the points he makes in chapter four.”

It’s easy to see. Dawkins thinks that God must be complex to do what he does, and he links complexity to improbability – since complexity involves numerous parts which are arranged in a statistically improbable manner. A God would be highly complex, so a God is statistically very improbable, and so God almost certainly doesn’t exist.
It’s all there in premise 3. A bad way to write an argument, yes – he combines a refutation of theism with a defense against the cosmological argument, and rather takes for granted the complexity of God – but it does support the conclusion that God almost certainly doesn’t exist.

  (Quote)

W January 7, 2010 at 10:48 pm

A ‘true’ account? Being a supernatural explaination is enough. Ok, Dawkins should have given more support other than that the explaination is itself unexplained. His point SHOULD have been that this ‘explaination’ is a non-explaination, every supernatural explaination that have been offered has failed. That’s neither here nor there, we can come up with a dozen of ways that.he could have rephrased it, and maybe that IS to do with his lack of philosophical training. In the end though, what does this matter? A charge of ignorance of philosophy of science to a scientist who has made considerable contributions to his field, seems pretty empty.

  (Quote)

lukeprog January 7, 2010 at 11:00 pm

Silver Bullet,

Did you just say that the explanation for everything surely requires an explanation, or did I misread you?

  (Quote)

lukeprog January 7, 2010 at 11:02 pm

W,

No.

Please read Dawkins’ quote. Dawkins does NOT say “To explain [something] by invoking a supernatural Designer is to explain precisely nothing, for the supernatural has no explanatory merit.”

Dawkins DOES say “To explain [something] by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer.”

  (Quote)

lukeprog January 7, 2010 at 11:04 pm

drj,

Please read Dawkins’ quote. In that quote he says that God is a poor explanation precisely because it leaves God unexplained. And that is just wrong.

If you want to talk about a complexity argument, well, I am getting to that, but it is not the point I have made in these two posts so far.

  (Quote)

lukeprog January 7, 2010 at 11:12 pm

TaiChi,

If you can produce a coherent formulation of Dawkins’ argument, I would greatly appreciate it. I’ll shortly be assessing Wielenberg’s attempt at reconstruction, along with any others I can find.

I agree that Dawkins may be more interested with his points about complexity than his point about “…for it leaves unexplained the Origin of the Designer.” In fact, that end of the paragraph appears to be completely separate from the rest of it:

To explain the origin of the DNA/protein machine by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer. You have to say something like ‘God was always there’, and if you allow yourself that kind of lazy way out, you might as well just say ‘DNA was always there’, or ‘Life was always there’, and be done with it.”

Here, Dawkins makes exactly the argument I attribute to him, and it has no logical connection to his complexity argument that I can see. Dawkins says that explaining DNA with a supernatural Designer fails because it leaves the Designer himself unexplained. (That’s the part to which I object.) Dawkins reinforces that this is the point he’s making by saying that ‘God was always there’ is a feeble response, as if an explanation of the explanation is required.

Are we still in disagreement?

  (Quote)

lukeprog January 7, 2010 at 11:16 pm

W,

I didn’t say Dawkins’ biology was naive. I said one of his arguments in philosophy of religion (in which he obviously has no training) was naive.

  (Quote)

lukeprog January 7, 2010 at 11:36 pm

TaiChi,

You’re right about seeing how Dawkins could conclude that god almost certainly doesn’t exist. My bad. I deleted that part of my post, now. Thanks!

  (Quote)

Ajay January 7, 2010 at 11:53 pm

Luke,

Isnt “dark matter” just a place holder? I dont think any scientist speculates or says with certainty what that dark matter is made up of or what it does. I dont think they mean “dark matter” as an explanation. I think it is just more of a “i dont know” and let us call it as “dark matter” for now.

But in the case of design of the universe and God, it is more like… “Eureka… i know the answer for the design of the universe. It is very simple. It is God. We have solved the mystery of how everything in this world came about”.

So, the only reason they say God exists is to simply solve the problem of the origin of the universe. But since that explanation of God sucks and it doesnt solve anything except raise more questions, we can conclude that God almost certainly doesnt exist. However, if we have other reasons to think that God exists like our prayers working everytime or miracles happening, then apart from the explanation of the universe we have other reasons to maybe believe in God and therefore we cant come to that conclusion.

So, the key here for me is whether we are invoking the God explanation just to explain the origin of the universe and after debunking all other miracle claims etc, we can also say that God almost certainly doesnt exist.

  (Quote)

W January 8, 2010 at 1:22 am

Isn’t Craig’s biology naïve? For him to even participate in a debate over intelligent design shows this, yet you defend this! I tell fellow atheists all the time that they need to brush up on their philosophy, I agree that Dawkins is not philosophically sophisticated. But if this means Dawkins needs a philosophy class, Craig needs a biology class. This defense of Craig has gone a bit far, maybe you are trying to hard to give hom the benefit of the doubt, I don’t know. I do love you site and agree with about 95 percent of what you say, but it seems that in trying to be fair with theists, you’ll let some things slip. Yes, Craig is right, you don’t need an explainationm for the best explaination. Dawkins should have consulted some atheistic philosopher, I really don’t know why he didn’t. But this amounts to what? He should have worded it better, big deal. For Craig to ignore the supernatural part and foucs on the best explaination is obvious sophistry, the best explaination has never been supernatural.

  (Quote)

TaiChi January 8, 2010 at 1:22 am

Luke,
“In fact, that end of the paragraph appears to be completely separate from the rest of it.”

It doesn’t appear that way to me. First: in both TBW and TGD Dawkins is *specific* in saying that theism has a problem with explaining “who designed the designer”, and in neither does he cite the general principle which you and Craig read into him, that any explanation must be itself explained. That suggests (and you should grant it as a matter of interpretive charity) that Dawkins is making a local claim. And as a local claim, it is entirely correct – the God hypothesis suffers from the special difficulty that what it proposes is less plausible than just accepting the existence of our complex universe as a brute fact. I’m sure you’d agree with me that if any explanation is to be adequate, it has to be better than no explanation at all, and the God hypothesis fails the test, as Dawkins points out.

Second: the quote you take out directly follows from, and is part of a paragraph with, Dawkins’ point that any God would have to be complex. It would be reasonable to suppose, then, that these ideas go together… however your reading voice happens to splice it in your head.

I’ll get back to you on the formal argument thing, once I’ve reread the chapter. There are probably three separate arguments I can tease out.

  (Quote)

W January 8, 2010 at 1:54 am

Sorry for the typos and sloppy spacing, I’m on a blackberry.

  (Quote)

Shane January 8, 2010 at 4:17 am

Luke, sorry, but you are mixing up extrapolation and interpolation. The humans designing the arrowhead do not need explained as part of the explanation for the arrowhead BECAUSE they are a “given” in this system. There is no such “given” for supernatural designers.

In this respect, Dawkins’ argument is perfectly sound, and Craig’s (and your) criticisms are, while interestng, invalid. You are changing the meaning of “explanation” mid-track.

Longer post on http://answersingenes.blogspot.com :-)

  (Quote)

Silver Bullet January 8, 2010 at 7:17 am

Luke:

You did misread me, but I didn’t make myself perfectly clear. Here’s what I meant:

“It seems to me that as an explanation for the origin of everything, the god argument does fail exactly because it simply begs the question of the origin of god.

Thus, while Craig may be correct when suggesting that an “ordinary” explanation should not be discarded because that argument itself requires an explanation, the explanation for everything surely must” be discarded for this reason.

  (Quote)

lukeprog January 8, 2010 at 7:30 am

W,

Then I think we agree. Craig certainly engages in some sophistry.

  (Quote)

lukeprog January 8, 2010 at 7:33 am

TaiChi,

I appreciate your thoughtful comments on this, but I’m just reading Dawkins differently. He specifically says that offering God as an explanation fails BECAUSE it leaves God unexplained. He could have easily written that offering God as an explanation fails because God is more complex than that which he is explaining. In fact, he does that elsewhere in his work. But in this instance he says something different – that God is a poor explanation because it leaves God unexplained. I’ve also heard him say exactly this – completely divorced from his complexity argument – during interviews. It is, after all, one of the most popular atheist rebuttals: “Who designed the designer?”

  (Quote)

lukeprog January 8, 2010 at 7:34 am

Shane,

I’m confused. Do you agree that in order for X to be the best explanation of Y, we need not also have an explanation of X?

  (Quote)

lukeprog January 8, 2010 at 7:36 am

Silver Bullet,

So if atheists ultimately arrive at a Theory of Everything similar to string theory, will it destroy the whole theory to say that we do not have an explanation for the strings themselves?

  (Quote)

Steven Carr January 8, 2010 at 7:57 am

CARR
According to Craig, if we saw machinery on the Moon, we would naturally assume that no God designed this machinery.

Why? Why does Craig suddenly rule out God as an explanation?

Where did Craig’s God go?

Is it because the circular nature of his arguments would immediately be exposed if he claimed that we should assume his (alleged) God designed something, simply because we don’t know the origin of something.

  (Quote)

Steven Carr January 8, 2010 at 8:03 am

Is Dawkins logic circular?

It is like asking who fathered the Father.

According to Christian doctrine, the Son had a Father, but , by definition, the Father had no Father, although the Son and the Father are One Being.

So Dawkins question is like asking who fathered the Father.

  (Quote)

MauricXe January 8, 2010 at 8:29 am

TaiChi: Second: the quote you take out directly follows from, and is part of a paragraph with, Dawkins’ point that any God would have to be complex. It would be reasonable to suppose, then, that these ideas go together… however your reading voice happens to splice it in your head.

This is probably the crux of some of our arguments.

  (Quote)

lukeprog January 8, 2010 at 8:43 am

Or, see the end of The God Delusion, chapter 3:

[My] whole argument turns on the familiar question ‘Who made God?’… A designer God cannot be used to explain organized complexity because any God capable of designing anything would have to be complex enough to demand the same kind of explanation in his own right. God presents an infinite regress from which he cannot help us escape.

Here again Dawkins seems to be associating his argument with a requirement that a successful explanation itself be explained. Have I misread Dawkins?

At the very least Dawkins implies that each step back in explanation toward the supposed Ultimate Explanation must be simpler, not more complex… which is also false. We regularly offer successful explanations that are not themselves explained, and we even offer successful explanations that are somewhat more complex than the things they explain.

  (Quote)

drj January 8, 2010 at 9:07 am

Once again, I think most of this I think can be chalked up to Dawkins’ imprecision and ambiguity. As a result, most of us are arguing about what we think he means, rather than what he said.

But in any case, I don’t see his argument making claims like this at all:

“All explanations themselves require explanations”

So I think Craig is guilty of a straw-man. Craig would be right, if his representation was accurate – that this would invite an infinite regress. But an infinite regress is exactly the type of absurdity Dawkins is trying to shove upon the theist through his argument. Dawkins seems to be trying to say something more like:

“Complex designers themselves require explanations”.

And of course, “complex designer” is not code for “all explanations”.

The complex designer requires an explanation, and by theist logic this must be another complex designer. The complex designer of the complex designer of the complex designer would also require an explanation, ad infinitum.

I don’t see anything particularly wrong with that, and one can see exactly why he makes this argument, given the nature of Intelligent Design. The logical way out for the theist is to argue divine simplicity.

  (Quote)

MauricXe January 8, 2010 at 9:09 am

lukeprog: Here again Dawkins seems to be associating his argument with a requirement that a successful explanation itself be explained. Have I misread Dawkins?

Yes. In fact, he makes it clear in the section you quoted; any argument that claims complexity needs a designer, is self defeating because it leads to infinite regress. Why? Because a designer is complex, if not more complex than his creation. Because a designer is complex, his complexity needs a designer.

lukeprog: At the very least Dawkins implies that each step back in explanation toward the supposed Ultimate Explanation must be simpler, not more complex… which is also false. We regularly offer successful explanations that are not themselves explained, and we even offer successful explanations that are somewhat more complex than the things they explain.

I think this would be a more accurate, or at least more on track, rebuttal of his argument.

  (Quote)

drj January 8, 2010 at 9:34 am

lukeprog: At the very least Dawkins implies that each step back in explanation toward the supposed Ultimate Explanation must be simpler, not more complex… which is also false. We regularly offer successful explanations that are not themselves explained, and we even offer successful explanations that are somewhat more complex than the things they explain.

I might have to re-read some stuff, but I don’t even know if Dawkins really implies that sort of thing either. Remember, he is borrowing theistic intelligent design logic to try to demonstrate that when one applies it consistently, it leads to infinite regress of designers. I don’t know if he actually believes all explanations must always be simpler than the thing they explain.

I do believe I read an answer that he gave in some questionnaire, where he said he suspects that all complexity in existence arose from simplicity, in an evolution-like manner. However,I don’t think this entails that things always evolve to be more complex or that all explanations will be simpler than that which they explain.

  (Quote)

Paul January 8, 2010 at 9:40 am

lukeprog: Or, see the end of The God Delusion, chapter 3:

[My] whole argument turns on the familiar question ‘Who made God?’… A designer God cannot be used to explain organized complexity because any God capable of designing anything would have to be complex enough to demand the same kind of explanation in his own right. God presents an infinite regress from which he cannot help us escape.

Here again Dawkins seems to be associating his argument with a requirement that a successful explanation itself be explained. Have I misread Dawkins?

At the very least Dawkins implies that each step back in explanation toward the supposed Ultimate Explanation must be simpler, not more complex… which is also false. We regularly offer successful explanations that are not themselves explained, and we even offer successful explanations that are somewhat more complex than the things they explain.

I read this differently.

I read it roughly as follows. If we are trying to explain how complexity came to be then how could something complex have given rise to it? In other words complexity led to complexity and thus a real answer has not been provided. In other words he is refuting the answer because the answer is actually part of the (original) question.

It is not asking an explanation to the best explanation.

  (Quote)

Tony Hoffman January 8, 2010 at 9:42 am

Luke,

I agree with Silver bullet’s response to you, and I think SB is on the same tack as me so I won’t repeat.

Luke: Here again Dawkins seems to be associating his argument with a requirement that a successful explanation itself be explained. Have I misread Dawkins?

I believe you have, or at least you are not reading him as generously (or more rigorously) than perhaps his language demands. I understand the point that you are making that an explanation cannot be discounted because it invites an infinite regression; I don’t think any reasonable person, including Dawkins, would choose to disagree with your argument there.

I think that Dawkins was trying to express, however inelegantly, that the explanation for ultimate origins of “God” is simply ad hoc, and fails to answer the question any better than “Brute fact.” It seems to me that that is the correct (generous?) reading of Dawkins’ argument in that regard, and that criticizing it as Craig has seems like a straightforward case of strawmanning.

Luke: So if atheists ultimately arrive at a Theory of Everything similar to string theory, will it destroy the whole theory to say that we do not have an explanation for the strings themselves?

I think that the Theory of Everything is not a metaphysical theory, just a theory of everything physical. I wonder if there isn’t a certain theistic / philosophical bias in some of us that feels an explanation that doesn’t have an ultimate end is patently unsatisfying; I don’t feel this way, but I wonder if Craig, you, et al. have this tacit conviction?

  (Quote)

Silver Bullet January 8, 2010 at 9:50 am

lukeprog: Silver Bullet,So if atheists ultimately arrive at a Theory of Everything similar to string theory, will it destroy the whole theory to say that we do not have an explanation for the strings themselves?  

I don’t know very much about string theory specifically, but it seems to me that whatever the proposed explanation for everything is, there must be good reason to consider it a brute fact (or at least based on brute facts). It would not suffice to make the ad hoc assertion that the explanation is a brute fact, for the origin of that explanation would be begged, and one wouldn’t at all have an explanation for everything in that case.

I’m not sure what reasons could suffice to consider strings to be brute facts, partly because I just don’t know enough about string theory, and partly because this feels like a philosophical question without an immediately intuitive answer and that I haven’t really considered much. I’m open to the notion that one may never be able to arrive at sufficient reasons to believe that something is a brute fact. Even if that were the case, Dawkins’ comments about the creator-god would, I think, remain true: inferring the existence of a creator god, and then making the ad hoc assertion that this god is a brute fact, doesn’t seem to provide an explanation for everything either. Dawkins also attacks the inference of a creator god as an explanation for the complexity and improbability of the universe, as I have explained in a previous post.

  (Quote)

lukeprog January 8, 2010 at 12:58 pm

Silver Bullet,

Dawkins is correct that God is not a good explanation for everything. That is not what I’m disputing. I’m disputing the REASON Dawkins gives for asserting that God is not a good explanation. Dawkins says God is not a good explanation because (among other things) it leaves the explanation itself (God) unexplained. That is NOT the reason that God is a poor explanation.

  (Quote)

TaiChi January 8, 2010 at 1:50 pm

lukeprog: But in this instance he says something different – that God is a poor explanation because it leaves God unexplained. ”  

Well, I can’t see that we’ll resolve this point – if you wish to stick to your interpretation even though there is textual support for a charitable reading, then I can’t argue with you.
But let’s suppose you are right in claiming that Dawkins makes this little point, and that he is incorrect. In that case, Dawkins gives two arguments: the God hypothesis fails as an explanation of the universe, (1) because God is more complex than that which he is explaining, and (2) because the explanation, God, is left unexplained. Given that (2) is incorrect, you still have to deal with (1). So I suggest that we ignore (2) as a bad argument on your view and an inessential elaboration of (1) on mine.

lukeprog: At the very least Dawkins implies that each step back in explanation toward the supposed Ultimate Explanation must be simpler, not more complex… which is also false.

I think that’s an accurate read on Dawkins. Why is it false?

  (Quote)

TaiChi January 8, 2010 at 2:18 pm

Here’s the argument I promised you, Luke. Actually, it’s two arguments, one of which is a sub-argument, and a corollary.

***Dawkins’ Anti-Cosmological Argument***
1. The universe is complex, and so statistically improbable. It therefore requires explanation.
2. One explanation for the complexity is design.
3. But design is, in any case, not an acceptable explanation.
4. An alternative model is provided by natural selection.
5.The multiverse hypothesis, combined with the anthropic principle, suggests that a similar model is possible to explain the universe.
6. We should therefore eschew the design hypothesis (from 3), and tentatively lean toward the multiverse hypothesis, combined with the anthropic principle.

***Dawkins’ Anti-Design Argument (Support for 3)***
A. An acceptable explanation is an account of something improbable in terms of what is more probable.
B. The hypothesis that God designed the universe should therefore, if it is to be explanatory, account for the universe in terms of what is more probable than the universe.
C. But God himself would have to be more complex than the highly complex universe.
D. What is complex is improbable, and is improbable to the degree it is complex.
E. So, the hypothesis that God designed the universe is not an acceptable explanation.

***Dawkins’ Corollary of C and D***
C. God himself would have to be more complex than the highly complex universe.
D. What is complex is improbable, and is improbable to the degree it is complex.
#. God is highly improbable. That is, he almost certainly does not exist.

Hopefully I’ve kept Dawkins’ structure pretty much intact. The first argument is given at the end of Ch.4, but I’ve taken some of the verbiage out and added the implicit premise that the universe requires explanation. The second argument expands on Dawkins’ third premise, taking from Ch.4.
The corollary also draws on Ch.4, and its being a corollary rather than the conclusion of the argument explains the otherwise bizarre fact that it is not listed as point 7 in the text, and is not prefaced by ‘therefore’. As Dawkins writes “If the argument of this chapter is accepted, the factual premise of religion – the God hypothesis – is untenable. God almost certainly does not exist.” – that is, if the anti-cosmological argument and its anti-design sub-argument are accepted, then one can easily draw the conclusion that the God hypothesis is probably false.

That’s my take. I hope it helps.

  (Quote)

lukeprog January 8, 2010 at 2:22 pm

TaiChi,

Yup, I’m moving on to the complexity stuff my next post in the series.

Please don’t say “I can’t argue with you.” Many, many times I have changed my mind about something only after extended interactions with people who disagree with me. The “Jesus is not magic” escapade (see the archives) is a recent example. You’ve already shifted my mindset into thinking that “leaving the explanation unexplained” is a more minor point in Dawkins than I had earlier thought.

Cheers.

  (Quote)

Silver Bullet January 8, 2010 at 3:21 pm

lukeprog:
…Dawkins says God is not a good explanation because (among other things) it leaves the explanation itself (God) unexplained. That is NOT the reason that God is a poor explanation.  

But it absolutely seems to me that it is one of the reasons why the god explanation does fail.

The explanation to end all explanations (god) cannot be left unexplained precisely because it is meant to be the explanation to END all explanations.

In this respect, god, as an explanation for EVERYTHING, is different from all other explanations, which are the ones Craig is addressing. Other, “ordinary”, explanations cannot be used as an analogy for the explanation of the origin of everything, which is different in the very respect that you and Craig incorrectly criticize Dawkins’ argument.

To simply assert that god requires no explanation, ad hoc, as I believe Craig does, is insufficient.

Thus, god – as an explanation for everything – fails, and it fails precisely because there is nothing about that particular explanation that indicates that a further explanation is not required.

If I have this wrong, can you help me to see where I am going wrong?

  (Quote)

Shane January 8, 2010 at 3:26 pm

Luke, I don’t think you are following Dawkins’ argument. In this context, it is adequately clear that the “explanation of X” means “reason to believe that X exists”.

If you want to invoke God as an explanation for the universe, you need another “reason to believe that God exists”. An explanation, in other words. For the arrowheads, we KNOW that people exist; no further explanation of their existence is required.

Interpolation vs extrapolation.

Are you getting this yet? ;-)

  (Quote)

lukeprog January 8, 2010 at 4:00 pm

Silver Bullet,

I don’t think Craig said that God requires no explanation. And I don’t think Craig would say God is an explanation for everything, especially the properties God himself possesses. But remember, we aren’t considering any positive arguments from Craig. We are criticizing Dawkins assumption that God fails as an explanation because the explanation itself is unexplained. That is NOT why God fails as an explanation. God as an explanation fails because it has poor explanatory scope, no testability, low plausibility, does not fit with background knowledge, etc.

  (Quote)

lukeprog January 8, 2010 at 4:01 pm

Shane,

Sorry, I and most philosophers of science disagree with you. In order to posit quarks as a good explanation for certain phenomena, we do not need to also have an explanation for quarks. Science doesn’t work that way, and never has.

  (Quote)

Silver Bullet January 8, 2010 at 6:39 pm

Luke,

It seems to me that Craig does just assert that god is uncaused (ie. without prior explanation). He did not say this in the quote you provided in this thread, but he does seem to do so in, for example, Strobel’s book, The Case for a Creator. If he isn’t just making this shit up, then how can he know that god is uncaused? In any case, I can provide a quote if you like, but I actually think that this takes us off on a tangent – away from your and Craig’s criticism of part of Dawkins’ argument.

I don’t think that you have addressed the concerns I expanded upon in my last post regarding god as an explanation for everything:

1. Positing “goddidit” begs the question of the origin of god. There is no good reason to believe that god has no prior explanation or cause. One can assert that, but it is just ad hoc.

2. The explanation for the origin of everything CANNOT beg this question: there MUST be good reason(s) to believe that the explanation for the origin of everything has no prior explanation or cause (ie. is a brute fact).

3. Therefore, “godidit”, as an explanation for everything, fails, precisely because it leaves open the explanation for the origin of god.

I appreciate that there are other good arguments for why the god explanation fails, but what I want to know is : what’s wrong with the argument above?

  (Quote)

Tony Hoffman January 8, 2010 at 7:08 pm

Silver Bullet,

I just want to say that every time I feel like chiming back in you beat me to it with a better version of what I was about to say.

Luke,

I appreciate your continued engagement on this with the wide variety of criticism you have faced, and that you appear to be considering your prior position in light of what seems like a revolt among your admirers. I wonder this: do you have a non-ambiguous reference where Dawkins says that all explanations must themselves be explained? It seems like a fairly foolish thing to express, and I have sometimes found that Dawkins (while fully capable on his own, as we all are) sometimes seems to be unfairly attributed as the source for all things both asinine and atheistic.

  (Quote)

lukeprog January 8, 2010 at 7:25 pm

Silver Bullet,

1. Positing quarks begs the question of the origin of quarks. This does not mean the quarks do not successfully explain certain quantum phenomena.

2. Agreed.

3. No. See (1).

  (Quote)

lukeprog January 8, 2010 at 7:26 pm

Tony,

No, I doubt Dawkins has ever said any such thing, as there would be no reason to say it without context. But the unqualified “who made God?” retort he often uses assumes it. I’m glad you think it would be a foolish thing to suggest.

  (Quote)

Silver Bullet January 8, 2010 at 7:47 pm

Luke,

I’ll try one last time.

The explanation for certain quantum phenomena is nothing like the explanation for the origin of everything.

Similarly, the explanation for the arrowheads in Craig’s example is nothing like the explanation for the origin of everything.

These examples are of explanations for items within the set “everything”.

Quarks and people, the explanations for the examples provided by you and Craig, would be expected to have explanations of their own origins, as would all items in the set “everything”. There is no reason why this expectation should disqualify these explanations.

On the other hand, the explanation for the origin of the entire set “everything” would be expected to NOT have an explanation for its origin, since it is the explanation to end all explanations of origin, so to speak.

Since “godidit” begs the question of the origin of god, “godidit” must fail as the explanation for the origin of everything: there is no good reason to believe that god is a brute fact.

  (Quote)

lukeprog January 8, 2010 at 8:46 pm

Silver Bullet,

Why does “God did it” beg the question of the origin of god, but, say, a multiverse does not beg the question of the origin of the multiverse, or a web of vibrating strings does not beg the question of the origin of a web of vibrating strings?

  (Quote)

Silver Bullet January 8, 2010 at 9:39 pm

Luke,

String theory seems to help reconcile quantum theory (the theory of the very small – sub atomic particle scale) with general relativity (the theory of the very large – universe scale) to produce a unified theory of gravitation. To my knowledge, String theory does not propose to explain the origin of the universe.

Similarly, the multiverse theory seems to explain some mysterious features or consequences of quantum theory. To my knowledge, the multiverse theory does not propose to explain the origin of the universe(s).

Perhaps I am wrong about the above, but if I am not, then I would ask why you keep bringing up examples of explanations for the origins of items in the set “everything”? Quarks, arrows, strings, multiverses…

The explanation for the origin of “everything” cannot itself have another explanation. That is the type of explanation I am referring to. Let’s please discuss this explanation, and not discuss lesser analogies.

You have previously agreed that this type of explanation must be a brute fact, or be based on brute facts.

There is no reason to believe that god is a brute fact. One could assert that god is a brute fact, but that would just be ad hoc. Therefore, it seems to me that god, as an explanation for the origin of everything, fails, precisely because the origin of god is begged.

(By the way: Dawkins’ main point is not the point I am working on above. I just re-read that chapter of the God Delusion; his main point is that the god-designer is a terrible explanation for the complexity and improbability of the universe because such a god-designer must be as or more complex and improbable as the universe whose complexity and improbability the god-designer is meant to explain. The only reason Dawkins even brings up the issue of the god-designer explanation leaving the origin of the god-designer unexplained is to force his readers to consider the improbability of the origin of the complex god-designer, which, in turn, forces one to recognize the self-refuting nature of the god-designer argument as an explanation for the complexity and improbability of the universe. Accordingly I think that you and Craig have focussed on a strawman. Nevertheless, as I have been trying to argue above, I think that even that strawman argument, which is an argument by analogy, fails because the explanation for the origin of everything is unlike the example explanations you and Craig have provided as described above)

  (Quote)

TaiChi January 8, 2010 at 11:16 pm

lukeprog: TaiChi,Yup, I’m moving on to the complexity stuff my next post in the series.Please don’t say “I can’t argue with you.”   

Sorry if that sounded a bit snippy – it wasn’t intended that way. I meant to state the fact that, having already made my case, there was nothing more I could say to convince you. All I could do was to repeat myself.

Looking forward to the next in the series.

  (Quote)

lukeprog January 8, 2010 at 11:52 pm

Silver Bullet,

Alas, I think we’re at a point of no progress…

  (Quote)

Shane January 8, 2010 at 11:57 pm

Luke, you’re still not getting where you have misunderstood Dawkins’ argument. Let’s try again.

lukeprog: Shane,Sorry, I and most philosophers of science disagree with you. In order to posit quarks as a good explanation for certain phenomena, we do not need to also have an explanation for quarks. Science doesn’t work that way, and never has.  

And I don’t think I said that. I also just throw in that you perhaps have too much respect for “philosophers of science” and not enough for scientists :-) Sometimes I would like to take some people who describe themselves as philosophers of science and actually get them into a lab, or get them to read the latest issue of Nature. The relationship between the “explanation” of quarks to the behaviour of fundamental particles is actually a *model*, and that model is testable. But that is not the point.

You have misrepresented Dawkins’ argument, not refuted, as you have claimed. At the very best you are talking at cross-purposes with D. Let’s try to re-shape things a little. You are right (trivially) when you say that for humans to be the explanation for the arrowhead, you do not need to explain the origin of humans. D would agree with that, because the explanatory SCOPE you need for the *arrowhead* has been sufficiently addressed. Remove the arrowhead, you are still left with humans, and explaining them is an entirely separate enterprise. There is no loose explanatory end FOR THE ARROWHEAD.

With this argument for the existence of the gods, things are quite different. If you invoke gods to explain the universe, you are left with an even bigger explanatory loose end than you started with, that actually completely encompasses the original loose end! You have explained nothing at all!

Now, perhaps oddly, Craig (whom I think you hold in too high regard, but never mind) might be RIGHT that you end up with a potentially infinite regress of explanations, but that is what science tackles – it opens ever more black boxes and tries to see how they inter-relate. None of this would be news to scientists.

However, when we open black boxes, we try to make the problems MORE tractable. By proposing ad hoc explanations like the gods, theists make the problems LESS tractable, and thereby increase the explanatory work required to provide an understanding.

You could think of it as the second law of explanationodynamics. The gods are counter explanotropic. Personally I think Dawkins covers this perfectly adequately, and Craig’s (and your) argument against Dawkins fails.

Nice try :-)
-Shane

  (Quote)

lukeprog January 9, 2010 at 12:16 am

Shane,

I probably do have more respect for scientists than philosophers of science. Philosophers are jealous of scientists for good reason. But it’s scientists themselves who continuously make successful explanations that are not themselves explained.

As I’ve said many times, I agree that positing God doesn’t help. But the reason it doesn’t help is NOT because God remains unexplained. It’s because God has poor explanatory scope, no testability, low plausibility, etc.

  (Quote)

Cafeeine January 9, 2010 at 12:39 am

I can’t help but feeling this is the most severe case of violent agreement I’ve encountered.

  (Quote)

josef johann January 9, 2010 at 3:12 am

You cannot require that explanations themselves be explained, for this leads to an infinite regress of explanations, and then we would never be able to explain anything.

This does not accurately relate Dawkins’ position.

TGD, pg 78

Some regresses do reach a natural terminator. Scientists used to wonder what would happen if you could dissect, say, gold into the smallest possible pieces. Why shouldn’t you cut one of those pieces in half and produce an even smaller smidgen of gold? The regress in this case is decisively terminated by the atom. The smallest possible piece of gold is a nucleus consisting of exactly seventy-nine protons and a slightly larger number of neutrons, attended by a swarm of seventy-nine electrons. If you ‘cut’ gold any further than the level of the single atom, whatever else you get it is not gold. The atom provides a natural terminator to the Crumboblious Cutlets type of regress. It is by no means clear that God provides a natural terminator to the regresses of Aquinas.

It looks to me like Dawkins is saying that some explanations resolve puzzles without requiring further explanation, (i.e. we can stop at the level of atoms and not fall into regress).

When he says..

To explain [something] by invoking a supernatural Designer is to explain precisely nothing, for it leaves unexplained the origin of the Designer

… he’s speaking in the context of an argument from improbability. In that context, an argument for a creator is self-defeating because it introduces more improbability than it explains. By itself, improbability doesn’t make an argument for a creator inherently wrong, it is only wrong when your premise is that improbability requires an explanation. That is a premise of all teleological arguments.

  (Quote)

lukeprog January 9, 2010 at 8:38 am

josef,

Page 78 says nothing about regresses of explanation. The issue there is about physical regress. Once you break down those particles past a certain point, it is no longer “gold” by definition.

In any case, I think it’s plausible that Dawkins was speaking only in the context of an argument from improbability, but I could swear I’ve heard him say “Who designed the designer?” before without the improbability context. I just can’t find the video right now. Anyway, please hold on to your hats for my next post on Dawkins, where things will change.

  (Quote)

MauricXe January 9, 2010 at 9:24 am

lukeprog: In any case, I think it’s plausible that Dawkins was speaking only in the context of an argument from improbability, but I could swear I’ve heard him say “Who designed the designer?” before without the improbability context. I just can’t find the video right now. Anyway, please hold on to your hats for my next post on Dawkins, where things will change.

He has. he has even said something of that sort in his first interview with Steven Colbert. I think his exact words where “And who did God?”

It is a misleading sound byte not to be taken for his full argument.

  (Quote)

Andrew Brenner January 9, 2010 at 9:52 am

Silver Bullet: The explanation for the origin of “everything” cannot itself have another explanation. That is the type of explanation I am referring to. … You have previously agreed that this type of explanation must be a brute fact, or be based on brute facts.There is no reason to believe that god is a brute fact. One could assert that god is a brute fact, but that would just be ad hoc. Therefore, it seems to me that god, as an explanation for the origin of everything, fails, precisely because the origin of god is begged.  

This isn’t at all relevant to Dawkin’s argument, but I just couldn’t help responding to Silver Bullet’s point here, as it seems to be based on significant confusion (or am I confused?). Silver Bullet claims that any “explanation for the origin of everything” must itself remain unexplained — it must be a brute fact about the world. Well, maybe this is so. But then the theist comes along and proposes God as the explanation for the origin of everything (or, more accurately, usually proposes that God is the creator of all contingent things aside from Himself). Silver Bullet retorts: “There is no reason to believe that god is a brute fact. One could assert that god is a brute fact, but that would just be ad hoc. Therefore, it seems to me that god, as an explanation for the origin of everything, fails, precisely because the origin of god is begged.” HOLD IT, didn’t Silver Bullet just say that any “explanation of everything” MUST be a brute fact, and didn’t the theist just claim that God in fact constitutes just such an explanation? Wouldn’t God then HAVE to be unexplained? How then does Silver Bullet get away with claiming the theist has made an ad hoc modification to their theistic hypothesis by claiming that the existence of God is a brute fact?

  (Quote)

lukeprog January 9, 2010 at 11:26 am

MauricXe,

Should it be taken as ONE of his arguments?

  (Quote)

josef johann January 9, 2010 at 11:37 am

lukeprog: josef,Page 78 says nothing about regresses of explanation. The issue there is about physical regress. Once you break down those particles past a certain point, it is no longer “gold” by definition.

Luke,

So too is an explanation sufficient or insufficient “by definition” depending on our requirements for a good explanation. Depending on those requirements an argument may or may not be said to fall into a regression.

Thus, the problem with the argument from impossibility, is with the requirements the argument is asking you to make of explanations.

Within the context of the AFI, it is only after you have judged other explanations to be insufficient that you can introduce a creator to resolve this “problem”. But it’s exactly this pivot which is at fault, because you can lean the exact same requirement against the postulated creator and cause just as much destruction.

So again, I don’t see that Dawkins demands that explanations themselves must be explained. He is showing this requirement is the undoing of a theist who chooses to adopt it.

And I don’t mean to be a contrarian, but it’s unclear to me what insight is to be purchased from the distinction between explanatory and physical regression. In the passage I cited, Dawkins’ operative word is neither ‘physics’ nor ‘explanation’ but ‘argument’. Since he juxtaposes the divisibility-of-gold against arguments for god, I think he regards them both as arguments. But he believes one argument doesn’t regress whilst the argument other does.

Perhaps I’m missing out on a difference between “argument” and “explanation”? Or maybe the point he is making with regard to Aquinas’ arguments in ch. 3 is different from the one he is making in ch. 4?

  (Quote)

Silver Bullet January 9, 2010 at 12:19 pm

Andrew Brenner:
How … does Silver Bullet get away with claiming the theist has made an ad hoc modification to their theistic hypothesis by claiming that the existence of God is a brute fact?  

If a scientific explanation for the origin of everything ever becomes available, there will be GOOD REASON(s) to consider that explanation a brute fact (or to be based on brute facts). There will not simply be the ad hoc CLAIM or ASSERTION that this explanation is a brute fact. I remain open to the possibility that such an explanation will never be found, but that has no impact on the ‘godidit’ explanation.

What is the good reason(s) to believe that god is a brute fact? Is there even one that does not result from circular reasoning (god simply MUST be a brute fact)? That is, is there even one that does not beg the question of the origin of god?

  (Quote)

Silver Bullet January 9, 2010 at 12:21 pm

josef johann:
So again, I don’t see that Dawkins demands that explanations themselves must be explained. He is showing this requirement is the undoing of a theist who chooses to adopt it.

I completely agree.

  (Quote)

MauricXe January 9, 2010 at 12:54 pm

No, not without an explanation of why he would say such a thing. I think what Silver Bullet says earlier is a good explanation:

Silver Bullet: The ID argument that Dawkins is addressing goes like this:

As he articulates in point # 1 above: the universe and life in it is too complex to have arisen out of chance, so it must have had an intelligent designer.

However, as Dawkins argues in point # 2 above, a creator god must be even more complex, and even less probable than the universe whose complexity and improbability the creator god is meant to explain. Accordingly, such an argument is self refuting – it can’t provide any explanation of what it is proposed to explain.

  (Quote)

Andrew Brenner January 9, 2010 at 4:35 pm

Silver Bullet:
If a scientific explanation for the origin of everything ever becomes available, there will be GOOD REASON(s) to consider that explanation a brute fact (or to be based on brute facts). There will not simply be the ad hoc CLAIM or ASSERTION that this explanation is a brute fact. I remain open to the possibility that such an explanation will never be found, but that has no impact on the ‘godidit’ explanation.What is the good reason(s) to believe that god is a brute fact? Is there even one that does not result from circular reasoning (god simply MUST be a brute fact)? That is, is there even one that does not beg the question of the origin of god?  

I was merely commenting on the idea that it would somehow be an ad hoc modification of the theist’s God hypothesis for him or her to claim that God was a brute fact. Considering that God is almost always offered as the “origin of everything” it is simply a part of that hypothesis that He would be a brute fact if, as you say, the explanation for the origin of everything must be a brute fact. You ask for a good reason to believe that God is a brute fact, when *by definition* God is the creator of everything besides himself (or all contingent things aside from himself) and would therefore, per *your* stipulation that the explanation of the origin of everything *must* be a brute fact, be a brute fact by definition. Your objection, then, really amounts to “why should we think God exists?” which I admit is a great question…

  (Quote)

Badger3k January 9, 2010 at 5:25 pm

lukeprog: Reginald,I think Craig’s illustrations for why you needn’t have an explanation of an explanation are perfectly useful. I gave further illustrations of my own. Seriously, this is a really basic principle of philosophy of science.  

I’ll never understand your love affair with Craig and his horrible arguments, but I think it boils down to this difference: you look at it through the lens of the philosophy of science, while we seem to be looking at it through the lens of actual science, as practiced by scientists. The standards are different, and sorry to philosophers, that’s what most of the world goes by. Dawkins was referring to the Argument from Design, not about where explanations come from. I’m sure you’ll say that I’m misunderstanding the argument, because it is always the other that is wrong. Right?

  (Quote)

lukeprog January 9, 2010 at 5:29 pm

Badger3k,

Appealing to actual scientists makes the problem worse. THEY are the ones who constantly offer successful explanations without the explanations themselves being explained. Philosophers of science learned that trick from scientists themselves.

  (Quote)

Antiplastic January 9, 2010 at 6:23 pm

Allow me first to add my “me too” voice to the chorus of people desperately trying to point out that if there is an interpretation of Dawkins arguments both plausible and obvious in which he does not make the asinine assertion that all explanations must have explanations, then it behooves people to go with that one.

Andrew Brenner
ehhh? That doesn’t seem true at all (not to mention it isn’t exactly relevant to the idea I took Smart to be advocating). In any case I can think of plenty of good empirical explanans that are more complex than the explanandum they are supposed to explain.

With equal predictive power?

Whether the explanans is a good one is, i would say, a factor of its ability to predict the explanandum (i.e. the probability of the explanandum given the explanans) and its prior probability. The prior probability of the explanans will no doubt be determined in part by its simplicity, but *only* in part. Once we concede that obvious point there’s simply no justification for saying a good explanans will *always* be simpler than the explanandum — there are other factors involved.

Of course there are other factors involved, predictive power being just one of them. We need to stop looking at the slogans people throw around at other people’s explanations and actually look at the mechanics of how they work.

Remember that parsimonious explanations are always going to be relative to the rest of your world model. Meaning, if the explanans and the bridge laws are already a part of your description, then claiming they explain the observation comes at no increase in the total complexity of your description of the world. (Note how this handily refutes your earlier assertion about Berklian idealism: materialism and idealism are observationally equivalent theories, but [idealism + Yahweh] requires additional bits to specify. You can’t just say “God made all the mental objects that we observe” and claim it’s simpler, because you still have to include “all that we observe” in your description!)

Example: I find a letter in my mailbox. I explain the letter by invoking highly complex human beings who wrote and delivered the letter. This is a good explanation in part because it has a high prior probability based on the fact that it meshes well with my background experience (every past letter in my mailbox was written and put there by a human being, let’s say).

Correct!

The complexity of the explanans (that is, the complexity of the physical/physiological makeup of the posited human beings – *not* complexity in terms of failing to mesh well with our background knowledge) is a non-issue.

Incorrect, because it contradicts the correct thing you just said.

It “meshes well with our background knowledge” just means that we already include letter-writing humans in our total world-model! Whereas letter-writing martians and letter-writing poltergeists are not included. But we could come to include them — we could come to accommodate the additional bits in the string it would take to specify them — if the predictive power of that hypothesis reaches some arbitrary level relative to the computational cost.

Hence my point in reminding people that even bible-believing christians who believe the KCA and are physicists don’t use Yahweh in their actual, real-world science. Yahweh doesn’t actually perform any real work in the model. So what you have is :

1) A complete description of all observations re: the earliest moments of the universe, and

2) EVERYTHING IN #1 plus a few sentences bolted-on at the end.

By definition 2 is less probable than 1, unless those precious few sentences entail the specific content of 1.

  (Quote)

Antiplastic January 9, 2010 at 6:32 pm

Andrew Brenner:
I was merely commenting on the idea that it would somehow be an ad hoc modification of the theist’s God hypothesis for him or her to claim that God was a brute fact. Considering that God is almost always offered as the “origin of everything” it is simply a part of that hypothesis that He would be a brute fact if, as you say, the explanation for the origin of everything must be a brute fact. You ask for a good reason to believe that God is a brute fact, when *by definition* God is the creator of everything besides himself (or all contingent things aside from himself) and would therefore, per *your* stipulation that the explanation of the origin of everything *must* be a brute fact, be a brute fact by definition.

You need a refresher course in what “ad hoc” means.

If I tell you about my cancer-curing pixie that cures cancer, and you ask me why on earth even if I do have a pixie anyone should believe she cures cancer, “but cancer-curing pixies *by* *definition* cure cancer” is not an argument.

This should go without saying, but “a definition” is not evidence. A stipulation is not evidence. A tradition of belief is not evidence. Only evidence is evidence. Everything else is ad hoc.

  (Quote)

lukeprog January 9, 2010 at 8:11 pm

Antiplastic,

“Who designed the designer” is such a common atheistic argument, especially among the New Atheists, I think it’s highly plausible that Dawkins has used that argument, especially since I swear I’ve heard him use in outside the context of his complexity argument, I just can’t friggin’ find it… :)

Also, take for evidence the huge number of atheist readers who commented here, insisting that “Who designed the designer” is a good comeback.

  (Quote)

Antiplastic January 9, 2010 at 9:23 pm

lukeprog: Antiplastic,“Who designed the designer” is such a common atheistic argument, especially among the New Atheists, I think it’s highly plausible that Dawkins has used that argument, especially since I swear I’ve heard him use in outside the context of his complexity argument, I just can’t friggin’ find it…
Also, take for evidence the huge number of atheist readers who commented here, insisting that “Who designed the designer” is a good comeback.  

But “who designed the designer” is an excellent comeback. A fatal one, which Dawkins employs to great effect.

An asinine comeback, which WLC either will not or cannot understand is not being made, is “if your explanation doesn’t come with an explanation I don’t have to accept your explanation.”

One merely asks of an explanation that it do what it says on the label, namely, explain what it claims it can explain. To say, as theists do, that organized complexity is a brute fact of reality is not to explain organized complexity, it is to say it has no explanation. It’s no more explaining anything than pushing your cold mashed potatoes into a crater around the plate is eating them.

  (Quote)

Andrew Brenner January 9, 2010 at 10:14 pm

Antiplastic,
So I wrote: “Example: I find a letter in my mailbox. I explain the letter by invoking highly complex human beings who wrote and delivered the letter. This is a good explanation in part because it has a high prior probability based on the fact that it meshes well with my background experience (every past letter in my mailbox was written and put there by a human being, let’s say).” You agree with this. I proceeded to assert that “The complexity of the explanans (that is, the complexity of the physical/physiological makeup of the posited human beings – *not* complexity in terms of failing to mesh well with our background knowledge) is a non-issue.” You disagree with this, and you feel it contradicts the first thing I said. I’m not sure why you feel I’ve contradicted myself. Physical complexity and failure to mesh with my background knowledge are two entirely different things… which you seem to conflate. In any case I think you’ve conceded my point (if there was any actual disagreement to begin with) concerning whether an explanans will always be simpler than the explanandum it purports to explain.

  (Quote)

Andrew Brenner January 9, 2010 at 10:23 pm

Antiplastic,
Let us assume that the explanation for the origin of everything, if there be one, must be a brute fact. God is (usually) by definition the creator of everything besides himself. It then follows that God, if He exists, is a brute fact. What is ad hoc about this? The original stipulation, that the explanation of the origin of everything must be a brute fact, was not made to render God a brute fact. The further stipulation that God is the creator of everything besides himself was also not made to render God a brute fact (for many it is simply part of the definition of “God,” for others it follows from His omnipotence). What’s the problem then? Are you going to start picking out other necessary theistic attributes and call them ad hoc as well? Is the fact that God is not an orange ad hoc?

  (Quote)

Antiplastic January 10, 2010 at 9:04 am

Andrew Brenner: Antiplastic,
You agree with this. I proceeded to assert that “The complexity of the explanans (that is, the complexity of the physical/physiological makeup of the posited human beings – *not* complexity in terms of failing to mesh well with our background knowledge) is a non-issue.” You disagree with this, and you feel it contradicts the first thing I said. I’m not sure why you feel I’ve contradicted myself. Physical complexity and failure to mesh with my background knowledge are two entirely different things… which you seem to conflate.

On the contrary. We need to make sure that we are comparing the same strings in order for the comparison to be meaningful. (Remember again, in what follows we are assuming equal predictive power between two descriptions.)

“Expanations should be simpler than what they explain” is a handy shorthand slogan, but it does not mean, and Dawkins and Smart and any sane person do not take it to mean, that the length of the explanans-string must always be shorter than the length of the explanandum-string. I take it this is the cartoon you are criticizing.

What you need to attend to is not the absolute length of the explanans-string, but the relative length of the string that is your total world-model. Translate “meshes with my background knowledge” into this vocabulary: the explanans-string is already incorporated into your world-model i.e. your background knowledge. All you need to do is connect this background knowledge to this particular observation. So relative to your background knowledge, your explanation is (in quantitative terms) simpler. It’s simpler because the complexity has “already been paid for”.

Of course, your background knowledge might include “I know to a moral certainty that all humans died in a nuclear holocaust yesterday and my mailbox and I are in a sealed concrete bunker 20 miles below the surface of the earth”, in which case, “mailman” will once again represent an addition of complexity to your model.

If the Yahweh-description-string is already a part of your model and is doing predictive work for you, then you can try to leverage this preexisting complexity into an explanation of the origin of the cosmos (provided you ignore the part of the Yahweh-description that unambiguously says he created the cosmos 6000 years ago). But the Y-D-S has to earn its keep, and “a sense of the numinous and transcedent” and “consolation” and “a source of moral strength”, while admirable qualities, don’t count in the realm of parsimony.

  (Quote)

Antiplastic January 10, 2010 at 9:43 am

Andrew Brenner: Antiplastic,
Let us assume that the explanation for the origin of everything, if there be one, must be a brute fact.

I’ve been assuming it for purposes of this discussion, but I want my objection to be on record: there is no final vocabulary, and hence no such thing as “absolute” brute facts, only theory- and vocabulary-relative brute facts. When we talk about “properties of facts” we are talking about properties of our mental descriptions which grant them certain statuses.

God is (usually) by definition the creator of everything besides himself.

1,000% wrong.

Zeus was the son of Chronos and Rhea, and not a creator of everything besides himself. Brahma is the creator, but he is self-born from the lotus flower growing from the navel of Vishnu. Depending on which book you read, Yahweh was only one of the 70 children of El, and no matter what nomenclatural juggling you have to do to cast him in the part in Genesis 1, it appears that vast oceans of water already existed before he raised our flat earth out of it and covered it with a sky-dome. And then there’s the Marcionite view that has OT-god as a lesser demiurge who happened to create the physical cosmos but who was separate from and inferior to the Heavenly Father of whom Rabbi Jesus was an incarnation.

Now, as it happens, all these beliefs are false. But they are not “by definition” false. Unless you tack on one additional arbitrary ad hoc quality to your concept. Which you do.

It then follows that God, if He exists, is a brute fact. What is ad hoc about this?

Correction: it follows that if BrennerGod exists, then it is a brute fact. But how do you know, for example, it is a “He”, and how do you know it has the additional descriptive property of caring whether his pronouns are capitalized? We have ‘BrennerGod’ as a concept, but you have to admit we can also have ‘BrennerGod-Alpha’ as a concept which is identical in every descriptive particular except for the uncreated bit, and ‘BG-beta’ which is identical in every respect except it’s a she, and ‘BG-gamma’ which is identical in every respect except he has 12 brothers, and ‘BG-delta which is identical in every respect except he died in 1962 etc. etc. ad infinitum.

Just because you have a “definition” or “concept” of something of having properties 1,2,…n plus the property of being uncreated, it does not follow that anything with properties 1,2,…n has the property of being uncreated!

What’s the problem then? Are you going to start picking out other necessary theistic attributes and call them ad hoc as well?

“Necessary attributes”. There’s a laugh. Tell me please that you understand the difference between “I assert that x is necessary” and “I have good evidence that x is necessary.” If you do, you will understand that asserting something without evidence to “get yourself out of a jam” is the very definition of ad hoc special pleading. It would not be ad hoc, if for example, you could supply the slightest bit of evidence that Yahweh has this property.

Of course, all this is almost beside the point. Even if God “necessarily” has some property which takes some finite number of bits to specify, the length of that description still counts against the total parsimony cost of the hypothesis. It is still a legitimate reason to exclude the entity from the model if it’s not earning its keep.

  (Quote)

Andrew Brenner January 10, 2010 at 2:55 pm

Antiplastic,
I was assuming, of course, that we were using “God” in the traditional western sense as an unembodied person who is omnipotent, omniscient, perfectly good, etc (what Oppy has a tendency to call an “orthodoxly conceived monotheistic god”). When referring to this person it is often customary in the english language to refer to Him as if He were a male, to capitalize pronouns that refer to Him, etc. Of course, you knew this already. When I refer to “necessary attributes” of God I mean necessary de re, as in “there is an x such that x is God and x necessarily has such-and-such attributes,” as opposed to the de dicto (what you take me to be saying) “necessarily there is an x such that x is God and x has such-and-such attributes.” For example, being a person, God is such that if He exists it is necessarily the case that he is not an orange. I don’t mean to say that necessarily God exists and God is not an orange.

As for complexity, we’re simply referring to different things, as I’ve already said. I also take it that the way in which I understood Smart is an accurate representation of the quote I cited. Perhaps there are other parts of his article that would make one understand that quote in a way that isn’t obviously incorrect.

  (Quote)

Antiplastic January 10, 2010 at 6:40 pm

Andrew Brenner: Antiplastic,
I was assuming, of course, that we were using “God” in the traditional western sense as an unembodied person who is omnipotent, omniscient, perfectly good, etc (what Oppy has a tendency to call an “orthodoxly conceived monotheistic god”). When referring to this person it is often customary in the english language to refer to Him as if He were a male, to capitalize pronouns that refer to Him, etc. Of course, you knew this already.

Don’t be so modest. I’ve done my time in the internet trenches and in academic philosophy and the only person I *ever* met who believed in Philosophergod but not Biblegod was a visiting speaker who believed in Korangod. Absolutely everyone has something a little more specific in mind than “orthodoxly conceieved monotheos”. But maybe you’re the exception to the rule, and you really do use the masculine capitalised pronoun out of a pure and disinterested respect for standard notation.

Let’s not get distracted from the the basic point here: Just because you have a “definition” or “concept” of something of having properties 1,2,…n plus some property x, it does not follow that anything with properties 1,2,…n has the property of being x. Agree or disagree?

When I refer to “necessary attributes” of God I mean necessary de re, as in “there is an x such that x is God and x necessarily has such-and-such attributes,” as opposed to the de dicto (what you take me to be saying) “necessarily there is an x such that x is God and x has such-and-such attributes.” For example, being a person, God is such that if He exists it is necessarily the case that he is not an orange. I don’t mean to say that necessarily God exists and God is not an orange.

I am quite familiar with the distinction, and understood you to be using de re. That does not in the slightest affect the appropriateness of my laugh-reaction. You can’t just go around “defining” the properties of things into existence, even if the thing you’re predicating them at exists! For the third time: just because you have a “definition” or “concept” of something of having properties 1,2,…n plus some other property, it does not follow that anything with properties 1,2,…n has the that other property, modal or otherwise. Agree or disagree?

As for complexity, we’re simply referring to different things, as I’ve already said.

And as I’ve already said.

I am talking about a popular and very devastating argument against the CA. You are talking about a feeble strawman version of it. You can hardly blame me for trying to insist you talk about the argument I am actually making.

Complexity is just the length of the string it takes to describe something. “Explanations shouldn’t be more complex than the thing they explain” is not just some arbitrary slogan atheist epistemologists pulled out of thin air so they can continue their philandering homobortionist lifestyles. Adding very long strings to your total world model when they perform no predictive work is the very definition of a useless, unparsimonious pseudoexplanation. Keeping track of such a long string is all cost and no benefit compared to treating the string of your pre-existing world model as (provisionally) ultimate.

Whereas your characterisation of the argument is what, exactly? That Smart and Dawkins are mental midgets who don’t realize that “a person did it” can be a good explanation of a simple observation, like a circle drawn in chalk?

  (Quote)

Andrew Brenner January 10, 2010 at 7:15 pm

Antiplastic,
“just because you have a ‘definition’ or ‘concept’ of something of having properties 1,2,…n plus some other property, it does not follow that anything with properties 1,2,…n has the that other property, modal or otherwise.”
I’m referring to a situation in which (so it seems to me) the property in question necessarily *follows* from other standard properties. So, for example, I might observe that from the fact that an orange is a physical object it necessarily follows that that orange is not identical with the number 7. This further property, “not-identical with the number 7,” isn’t a further property that’s been tacked on for secondary reasons, but follows inescapably from the orange’s other essential properties. I (and probably most philosophers of religion) claim that God having no explanation outside Himself (or no explanation simpliciter) *follows* necessarily from other central properties normally ascribed to God. This property hasn’t been tacked on for other reasons, and it isn’t an ad hoc modification made to the hypothesis to save it from refutation. Now, either I’m correct that this property follows from God’s other central properties, or I’m not. In any case, it should be undeniable that the addition of this property to our conception of God, if made for the reasons I specified earlier (that it follows necessarily from some of God’s other properties), is not an ad hoc modification to God’s properties, but a genuine (attempted) elucidation of what properties would follow from our standard conception of God.

“Whereas your characterisation of the argument is what, exactly? That Smart and Dawkins are mental midgets who don’t realize that ‘a person did it’ can be a good explanation of a simple observation, like a circle drawn in chalk?”
I wasn’t referring to Dawkins. But in any case, if Smart just wanted to tell us to mind Occam’s razor he could have done so in a far less misleading way. I was taking Smart to be offering a refutation of some sort of argument from design. If he really just wanted to say that God was an unnecessary addition to our ontology, one that wasn’t pulling any explanatory weight, than he could have been less misleading. Or, perhaps he wasn’t misleading and I’ve only misunderstood him. If, however, I have misunderstood him, and he should be understood in the way you’ve suggested (mind the razor!), then his argument becomes far less interesting.

  (Quote)

antiplastic January 17, 2010 at 6:51 pm

I doubt anyone is still following the thread, but just as an exercise:

Andrew Brenner: Antiplastic,
“just because you have a ‘definition’ or ‘concept’ of something of having properties 1,2,…n plus some other property, it does not follow that anything with properties 1,2,…n has the that other property, modal or otherwise.”
I’m referring to a situation in which (so it seems to me) the property in question necessarily *follows* from other standard properties.

I’m sorry, but I rather clearly asked you over and over again whether you disagreed with the proposition. I don’t see why I should bother asking a 4th or 12th or 300th time. I have to take this silence as acquiescence. So we agree: “just because you have a ‘definition’ or ‘concept’ of something of having properties 1,2,…n plus some other property, it does not follow that anything with properties 1,2,…n has the that other property, modal or otherwise.”

Which is really rather fatal.

So, for example, I might observe that from the fact that an orange is a physical object it necessarily follows that that orange is not identical with the number 7. This further property, “not-identical with the number 7,” isn’t a further property that’s been tacked on for secondary reasons, but follows inescapably from the orange’s other essential properties. I (and probably most philosophers of religion) claim that God having no explanation outside Himself (or no explanation simpliciter) *follows* necessarily from other central properties normally ascribed to God.

From which non-modal properties of Biblegod do Her modal properties follow? None, of course. Is it the obsession with penis-shape? The geocentrism?

This property hasn’t been tacked on for other reasons, and it isn’t an ad hoc modification made to the hypothesis to save it from refutation.

Hardly. The property is tacked on for what reasons specifically?

Now, either I’m correct that this property follows from God’s other central properties, or I’m not. In any case, it should be undeniable that the addition of this property to our conception of God, if made for the reasons I specified earlier (that it follows necessarily from some of God’s other properties), is not an ad hoc modification to God’s properties, but a genuine (attempted) elucidation of what properties would follow from our standard conception of God.

Good gracious, is that what you think this is about?

Whether green-ness is an ad hoc modification to the hypothesis of green gremlins who are the cause of all automobile problems?

No, greenness is not an ad hoc modification of the green gremlin hypothesis, and necessariness is not an ad hoc modification of the necessary biblegod hypothesis.

The question is, is necessity a modification of the hypothesis driven by any actual evidence you have? That question is rhetorical by the way.

“Whereas your characterisation of the argument is what, exactly? That Smart and Dawkins are mental midgets who don’t realize that ‘a person did it’ can be a good explanation of a simple observation, like a circle drawn in chalk?”
I wasn’t referring to Dawkins. But in any case, if Smart just wanted to tell us to mind Occam’s razor he could have done so in a far less misleading way. I was taking Smart to be offering a refutation of some sort of argument from design.

A pretty fatal one.

“One shouldn’t add bits to one’s world model if they perform no predictive work.”

Crazy, right?

  (Quote)

Gregory Abucks March 5, 2010 at 10:29 pm

“While this may be the ideal, this is not how science works. Good chemical explanations were offered for things before we knew anything about “atoms.” Today, scientists posit non-baryonic matter (”dark matter“) to explain certain observed features of galaxy rotation and clustering, even though they have no explanation whatsoever for the dark matter itself. And psychologists regularly explain behavior in terms of theoretical structures for which the underlying neuroscience is not the least bit understood.”

WE do not know why gravity works, but we know an explanation exists.

The point is that an explanation for God MUST exist, not that we can know it. In fact, what that explanation is is largely irrelvant. You are confusing practical and ideal. If God exists, then he is unknowable to us. Therefore, any knowledge of him would be an ideal that we cannot reach. Full explanatory knowledge is also an ideal, as you admit. When talking about God, where no real knowledge exists, we are only concerned with ideals.

And, again, the importance is simply that an explanation exists. Since a God hypothesis only has value as a “first cause” argument, this shows that such a hypothesis is largely extraneous if not false.

_______

“The obvious problem here is that Dawkins’ argument is totally invalid. No laws of logic would allow you to deduce that conclusion for those premises.”

Yes, well Dawkins was never big on the whole “logical progression” thing. What he was trying to make was the perfectly valid point that not knowing an answer doesn’t mean there isn’t one or that we should make one up.

  (Quote)

Leave a Comment

{ 2 trackbacks }