Tuesday, November 13, 2012

Doctor Logic, Placeholder Fallacy, and the football game

A redated post.

I am linking to Doctor Logic's presentation of the Placeholder Fallacy.

This is a very critical issue in the debate between theists and atheists. Theists point out the explanatory difficulties in materialistic explanations. Atheists acknowledge these problems, but then point out that bringing theism, or a soul, or something not materialistically helpful isn't going to help. They contend that we're stuck with materialism, because to offer, say, a theistic explanation is not to explain anything at all, but rather to provide a placeholder for the real explanation.

Here's the problem I see with this kind of argument. I watch Sunday Night football with church friends most weeks. One of them, Butch, is as Dallas Cowboys fan. I know a number of things about Butch that enable me to converse with him. I know that he is an English speaker. I know that he is a Christian. I know he likes the Cowboys, so if Romo throws a touchdown pass, I know he'll be pleased, and if he throws a pick-6, I know he won't be happy. I know that he acts for reasons. Now, since I happen to be a dualist, I think that these teleological explanations are basic explanations and these explanations of his behavior can't be reduced to neurophysiological explanation. But even if I were a materialist, it seems to me just insane to say that the explanations that I employ in order to form expectations about what Butch will say and do are just placeholders for mechanistic explanations that might be supplied by brain science, or for the explanations that might be provided at the level of basic physics. These are real explanations, and if dualism is true, I can still use them.

But what about God. After all, God's not an embodied being the way Butch is. But people do seem to know what they are praying for when the pray, for example. It stands to reason that a God who is infinitely intelligent should want there to be other intelligent beings in existence. I may not get much detail in my predictions about God's conduct, but I can form probabilistic expectations concerning what God can be expected to do. I may see through a glass darkly, but I'm not completely blind. I simply do not see that "God raised Jesus from the dead to vindicate his ministry and show him to be God's own Son" is a placeholder. Of course it's a false explanation if Jesus never rose from the dead, but it isn't a placeholder.


Steven Carr said...

'I may not get much detail in my predictions about God's conduct, but I can form probabilistic expectations concerning what God can be expected to do.'

So do you think your god would save a child burning to death in a blazing house?

How would you assess the probabililty of your god saving a child being gassed to death?

One in a million?

Gordon Knight said...

I don't think you need to have predictions about what God will do to argue that God is a plausible explanation for the existence of the universe. I see agent causation as providing the best model for understanding causation of the universe and agent causation requires, well, an agent.

We even have an analogy of this sort of causation in our own case, when we think of, create intentional objects.

materialism, being materialism, cannot allow for such agent causation.

So theism wins!

Steven Carr said...

Gordon describes deism and claims it is theism....

Gordon Knight said...

I said you don't need to make predictions about what God would do, not that God never interacted with the world, which I guess is what deism is (why is this not a kind of this theism, since there is still a Theos?, just a rather aloof one)

But I take it teh first thing is to prove God, then we can go on to quibble about how and to what extent God intervenes in the world

In other words, God as agent cause of the world is compatible with a more robust Christian God, as well as deisms and other theisms.

Mike Darus said...

Steven asked:

So do you think your god would save a child burning to death in a blazing house?

Mike (apparently stumbling into the trap) suggests:
It is more likely God sent a nice fireman since he has (in writing) delegated this type of responsibility (and opportunity to do good) to created beings. If He did it all Himself, He would rob us all of the opportunity. If He did so act, He would likely need to cloak his action so we wouln't get the misimpression that we are now off the hook.

Alan Rhoda said...

We need to distinguish between "mere" placeholders and "partial" placeholders. The former are simply labels for as yet undiscovered causes with no underlying theory or prospect of one. The latter include a more-or-less worked out sketch of a possible cause, including some of its explanatorily relevant properties and/or powers and liabilities. Moreover, partial placeholders may admit of further development. In principle at least, they can be made less partial by filling in more of the details.

It seems to me that it is only the use of "mere" placeholders that merits the charge of fallacy. It would be absurd to suggest that no one should ever posit an explanatory entity until one has fully worked out all the relevant details, or that one be able to make highly precise quantifiable predictions. Partial placeholders figure prominently in both common sense and in science.

Now, theistic explanations of, for example, the apparent fine-tuning of the cosmos, are not "mere" placeholders. They have some content, e.g., they attribute properties like power and intelligence to the Designer. Moreover, theistic explanations admit of further development. Theodicies, for example, are attempts at filling out the theistic hypothesis in response to the problem of evil. Such accounts show that theism is not prediction-less. Contrary to Dr. Logic's claim that "the mind of God, is an explanation we do not (and can never) have", most theists would hold that we can and do have significant, albeit partial, insight into the mind of God.

Steven Carr said...

It is more likely God sent a nice fireman since he has (in writing) delegated this type of responsibility (and opportunity to do good) to created beings.

You mean Jesus was a created being?

And nobody has yet come up with writings from this alleged god.

How can an omni-benevolent being refrain from doing all benevolent acts?

Blip said...

How can an omni-benevolent being refrain from doing all benevolent acts?

How can Carr refrain from derailing all comments threads?

Timothy David said...


Steven Carr said...

'I may not get much detail in my predictions about God's conduct, but I can form probabilistic expectations concerning what God can be expected to do.'

Blip complains it is 'derailing the thread' to ask what an omni-benevelent being should do.

Clearly Blip simply doesn't like forming 'probabilistic expectations concerning what God can be expected to do'.

As he knows atheists think a loving God would not let children scream as they burn to death in a blazing house, or are gassed in Nazi death camps.

Such thoughts are 'derailing the thread', even if they are directly relevant to a post about what we can expect this god to do.

I guess I haven't learned how to do natural theology.

Step 1. Look at the world.
Atrribute all the nice and amazing things to God. All the other (nasty, evil) things are not caused by God.

Step 2. As you have defined a god as somebody who does all the nice and amazing things we see around us, examine your concept of God to form expectations of what we should expect to see.

9 times out of 10, your expectations will then br that this god will do all the nice and amazing things we see around us (but not the nasty , evil things we also see happening)

Step 3.
Look at the natural world and see if the nice things exist that led you to conclude that there was a god who did those nice things.

Step 4.
Declare that your expectations of God have been vindicated and so you now have evidence for God.

Step 5.
Never start from first principles and ask what an omni-benevolent being would do if the creatures he loved were suffering horribly.

That way lies madness....

Steven Carr said...

But even if I were a materialist, it seems to me just insane to say that the explanations that I employ in order to form expectations about what Butch will say and do are just placeholders for mechanistic explanations that might be supplied by brain science, or for the explanations that might be provided at the level of basic physics.


That might well be why Dr. Logic's blog posting does nothing even remotely analogous to that claim.

His post is more like attacking the place-holder claim that Butch like the Cowboys because God implants a love for the Cowboys in our heart, just like this alleged god implants morality in us.

Is it insane to say that the explanation 'Butch likes the Cowboys because they are God's team' is not a real explanation?

Steven Carr said...

As Victor is now traducing opponents views , making straw men out of them, and calling them insane after he has distorted them so badly, it is time to end commenting on somebody who can write articles like that.

I'm sure Victor will be pleased.

SteveK said...

I agree with you here Victor. In his blog post, Doctor Logic says

"This fallacy is most commonly used by theists when they claim that God can explain the existence of the universe, the fine-tuning of physical constants, or the origin of species."

There are many different ways of using the word 'explain'. Do you think Doctor Logic is making the mistake of lumping Aristotle’s four causes into one term, ‘explain’? God grounds all of reality and so he is the ultimate explanation for everything in reality, but not in the same way for all things.

philip m said...


As long as you are doing what I think you are doing, which is mentioning the argument from evil *in approval* of it, then you are disagreeing with doctor_logic. Doctor_logic is not saying that the evidence for God is bad. Doctor_Logic is saying we can't even *start* to offer evidence for or against God, because God is just an anyonmous theory, capable of explaining everything, and for that reason is incapable of explainging anything.

But you are saying that we don't expect the horrible suffering we see in the world if God exists - oh! So God *can't* explain everything, is that what you're saying now? But that is just what Doctor_logic is saying God can do, since the nature of the God hypothesis is such that He acts as an explanatory stamp we can put on anything and everything.

But now that we can't explain evil, apparently you think there is more content to the God hypothesis than doctor_logic presumes is the case.

So think you and Victor are very much in agreement on this point.

Doctor Logic said...


I can't believe I missed this post!

You didn't quite understand my argument. Your model of your friend Butch is natural by my standards, even if dualistic. I am totally A-OK with dualism if it provides a predictive model. However, if dualism is just an argument that we don't know a thing, then it's not an explanation.

God, in my book, might be natural. All God has to do to be natural is to have some regularities, just like your friend Butch. Physicality is not necessary.

Here's the thing that is simply wrong about Christian theism (and most other kinds). If you notice X, you can't explain X with nothing but Y = "That which causes X".

Y is not an explanation of X. It is a description of what some Z has to be in order to explain X.

For example, suppose X = "the death of Fred".

I haven't explained the death with Y = "that which causes the death of Fred".

If you ask "How did Fred die?" or "Why is Fred dead?", it makes no sense for me to reply "Because of that which causes Fred's death!"

I can come up with all sorts of paraphrasing of Y and still be guilty of the fallacy. Suppose all that happens is "fate". I can say "It was fate!" and be just as guilty of the placeholder fallacy. Saying Fred's death was fate explains nothing whatsoever.

The problem that theism faces is that "It was God's will" is equivalent to fate UNLESS you have some predictability/personality to God's will that makes it more than fate.

For example, if God kills blasphemers with lightning bolts at especially high rates, then "Fred was a blasphemer, and God killed him for it" would be a good explanation.

Yet, Christianity is locked very tightly into the placeholder fallacy. It's generally considered bad form to predict God or claim to know the will of God in predictive terms. This reduces God's will to another name for fate.

When you come to try to explain evolution or the Big Bang in terms of God's will, you really end up back at an empty statement that equates to "It was fate." It's simply not an explanation.

What you really mean, I think, is that if you knew the mind of God, God's will would be explanatory. And I agree! It's just that you don't know God's mind except by what happens after the fact. Of course, I can say I know fate too by what happens after the fact!

shiningwhiffle said...

Doctor Logic, how does your argument handle the case of explanatorily basic phenomena, where there is no explanation and no placeholder for an explanation? Example: why a probability-function collapses in one way instead of another?

Zach said...

Wow, I can see why this blog used to be popular. How did we get from excellent discussions with Gordon Knight and Doctor Logic, to the vacuous pablum supplied by the low-brow antics of Crude and Papalinton doing nothing but trying to out-insult one another? The decline is truly sad.

Doctor Logic said...

ShiningWhiffle, there are descriptions and then there are explanations. Inexplicable or brute facts/laws can be described, at best.

Let's suppose that you have a model of God that accurately predicts what he will do. That model will be an explanation for the things God is alleged to hanvtoninve done because it predicts what he did and predicts what he will do next. However, this God theory is not an explanation for God himself. God's nature and existence is a brute fact that can only be described, not explained.

It's just like fundamental laws of physics. Those laws are an explanation for the phenomena we see, not for themselves. String theory (if we had that fundamental theory!) is not an explanation for string theory, but for the various low energy physical laws we see (e.g., electromagnetism). If string theory is fundamental, it can only be described, not explained. To explain it would require some deeper explanation (which by our premise of its fundamental nature does not exist).

I've already explained what makes God a placeholder and not a true explanation. Calling God an explanation is like calling "unspecified physical laws" an explanation. Theories of God are not specific enough to make any predictions.

Zach said...

DoctorL: so do you see an asymmetry between explanation and deduction? E.g., let's say for argument that I can deduce that God exists: how is that different from an explanation of God's existence? Sorry if you have answered this above and I missed it.

shiningwhiffle said...

Doctor Logic,

Thanks for the clarification.


No, though it'd be an explanation of your belief in God's existence.

Consider the Kalam argument, which if sound shows the universe has a cause. Now consider the hypothetical case where I walk into my office and find a blender on my desk.

I know blenders don't spontaneously appear without a cause, so I could deduce through the same logic as Kalam that there was a cause for the blender being on my desk.

Does that explain anything?

Zach said...

Siningwhiffle asked:
Does that explain anything?

That is the question, no?

1. F=ma
2. Initial conditions wrt this mass are this way.
3. Therefore, this mass has such-and-such property.

In that case, I have explained why this mass has such-and-such property.

There are arguments such as:
1. Reality has such-and-such properties.
2. If reality has such-and-such properties, then God exists.
3. Therefore, God exists.
Which could explain God's existence, no? And if not, why not? What is the asymmetry?

Kalam argument could even be taken as explaining why there is a first cause. That is, why there is a God.

I realize this is a bit counterintuitive, but is there a precise way to demarcate deductions that explain versus deductions that are not explanations? This seems to be fuzzy, though I do not know the literature on the topic.

What is an explanation of X. It is a rational inference from premises to X (it could be statistical or deductive in nature). What is it about rational inferences that explain versus rational inferences that do not explain? Is there a fact of the matter, or is it a pragmatic subjective thing? E.g., does it reduce to 'Rational inference I explains X if we are interested in explaining X.'

shiningwhiffle said...


I think you're equivocating a little here.

"Y happened/exists/is the way it is because of X" is an explanation of the Y.

"We know Y happened/exists/is the way it is because of X" is not an explanation of Y.

Now, the reason we know something may also be the cause. Example: Suppose I know I got paid today because it's the day they pay me. Well, then I also got paid today because today's the day they pay me.

But in your example, "I know God exists because reality has such-and-such properties" is very much different from "God exists because reality has such-and-such-properties."

Zach said...

I'm not talking about knowledge, so don't see any equivocation. I am talking about arguments and inference. Are you saying that what distinguishes explanatory inference from nonexplanatory inference is that the former explicitly mentions knowledge? That seems strange, as my explanation from physics above didn't mention knowledge.

Doctor Logic said...

Zach, deduction tells us what must follow on pain of contradiction if our theory (i.e., our set of premises) is true.

Bayesian induction is the method by which we assess which theory (set of premises) is likely to be true based on evidence.

Deduction is concerned with validity, and induction with soundness.

I don't believe there are purely a priori arguments or purely a posteriori arguments. It's more like a spiral staircase of phenomenology, epistemology and ontology. For example, the cosmological and teleological arguments rely on premises that are inductive. They're only weakly a priori in the sense that the inductive premises seem to be pretty general. A lack of cause for an event is not a contradiction in itself, but it is intuitively peculiar.

Anyway, Bayesian inference is the way you find the best explanation. The next question would be whether a theory, T, is explanatory of X, whether or not T is actually the correct explanation of X. (Example: suicide and murder might both explain a dead body, even if one turns out to be a much better explanation than the other.)

Of course, what constitutes an explanation is partly a matter of definition. However, if X can be explained by a theory like

T: an unspecified physical laws caused X

then everything is trivially explained by T. No one I know would consider T an explanation of X. At best, it's a statement like "there exist predictive explanations of X that are physical."

Likewise, if I say

G: a deity, for unspecified reasons, planned and caused X

is not explanatory, but is more like the statement that theistic explanations for X exist.

Now, if you have an a priori argument for God (I am skeptical, of course), it seems like you would only get as far as saying that all explanations are theistic without actually providing any particular explanation.

Imagine, (contrary to fact) that I have an a priori argument the says the universe is physicalist. I still haven't explained dark energy by showing that the explanation for it must be physical. I've merely showed that any explanation I do find will be a physical one.

Victor Reppert said...

Zach: I am sorry you think the quality has gone down here. I suspect that this has to do with my attempts to address people in the Gnu camp, particularly on Loftus' blog, with the result that some of them started posting over here.

William said...

Before this talking past one another goes on too long, I suggest people define what they expect an explanation to do, and then explain why the existence of God does or does not do manage to do such work, because that is the issue that people are tripping over here...

Doctor Logic said...

William, I think my last comment explained the issue.

The mere existence of God is no more explanatory than the mere existence of unspecified physics.

An explanatory theory of God could exist, in principle, and it would be similar to an explanatory theory of physics, i.e., it would be a specific theory that actually predicts something non-trivial.

In my experience, theists flee from this requirement because they don't want to hang their faith on predictions that might be wrong (like predicting the end of the world last year).

A proper explanation is a predictive theory, even if the explanation is theistic and relies on folk concepts like love, goodness and evil as fundamental qualities.

shiningwhiffle said...


What I'm getting at is that surely "God exists because reality has such-and-such properties" has the explanation backwards.

Humans exist because of reality's properties. Dogs and cats and rocks exist because of reality's properties.

Does God exist because of reality's properties? Shouldn't that be the other way around?

Sorry if I was confusing.

Zach said...

Thanks all I will have to process this for a bit. This is outside my usual thinking volume.

Victor: that actually seems a good diagnosis!

Crude said...


Adorable, but I actually - unlike most people - ignore Linton 99% of the time. My last real exchange with him was exposing him (for the second time, no less) being a liar, a plagiarist and a fraud. You know, exactly what I accused him of being all this time.

Most of the time I'm content to simply discuss things, though if someone trash-talks, yes, I trash talk right back. I'll happily put the intellectual content of my comments here against your own, which have for a long time been little more than "GRR, Crude make me ANGRY! Must fight Crude, poorly!" and hastily written, poorly-thought-out assertions about philosophy of mind.

But hey, thanks for the compliment. Considering you can barely enter a thread without seething with hostility about me, it's clear you're going to need some therapy to get me out of your head - it's nice to know I can still have that effect. ;)