Monday, September 09, 2013

Theistic science

Prominent atheist Jeff Lowder recently did a post on theistic scientific explanations:
Due to pushback from some astute commenters, he quickly got in over his head, so it looks like he's changed the subject to a new post. I'm going to quote some of the feedback:
Neil Shenvi • 6 days ago
Your objections are interesting, and I'll have to chew on them for a bit. However, your main objection seems heavily predicated upon the idea that theism cannot be an 'explanation' at all unless it provides a mechanism. But can you explain (no pun intended) what you mean by a 'mechanism'? You seem to assume that any 'mechanism' must involve a naturalistic description of causation (hence your objection to the Antarctic markings example used by Meyer in his book). But, in the case of theism, such a mechanism is impossible since theistic intervention would necessitate non-natural causation.
Moreover, I would question whether a mechanism really is a necessary condition for any explanation. For example, in my own field of quantum mechanics, scientists will routinely invoke 'decoherence' or 'wavefunction collapse' as explanations for observed phenomena even when the mechanism for these processes is not known either empirically or even philosophically. Yet it hardly seems appropriate to dismiss these inferences as 'non-explanatory' simply because the mechanism of the interaction is unknown.

Neil Shenvi Jeffery Jay Lowder • 6 days ago
Thanks for answering. I'm stuck on the following remark:
"When I refer to "mechanism," I'm simply referring to a description of how the cause "did it.""
Since you're open to the possibility of supernatural causation, could explain what a 'how' description would look like in the case of supernatural causation? All of the 'how' examples that you list appear to be mechanistic descriptions of naturalistic chains of events. Take the origin of the universe, which theists generally take to be creation ex nihilo. Could you explain what a 'mechanism' would look like for the supernatural causation of Nature from non-existence?
"Not only is that mysterious by itself, but when combined with our ignorance about how the Designer designed life, it becomes even more mysterious."
I'm not sure this line of reasoning is valid. Even if we granted that we've only observed minds associated with physical brains, design arguments do not therefore depend on 'mystification.' For instance, we could equally argue that we've never observed a sentient mind 'not dependent upon a _human_ brain' or that we've never observed a sentient mind 'not dependent upon a carbon-based body.' But would we be forced to conclude that SETI is doomed to failure because any explanation it provides will merely 'mystify' the problem of the origin of extraterrestrial signals? Obviously not.
Or imagine that one day we discovered repeatable evidence that Ouija boards allowed real communication with some unknown, intelligent source. Would we stare at the reams of generated information and still say 'Well, claiming that some disembodied mind is the source of this communication is just mystifying the problem and is not an explanation at all'? It seems to me that an open-minded person, even if they lean towards physicalism, would have to grant that -in principle- there could be evidence that would lead us to infer the existence of disembodied minds. But if that is the case, we can't reject the very idea of a divine Designer as 'mystification'.

Neil Shenvi Jeffery Jay Lowder • 5 days ago
Hi Jeffrey,
Correct me if I'm wrong, but your proposed solution doesn't seem to address the actual mechanism of supernatural causation. What you've provided is a description of a supernatural chain of events (side note: which is oddly similar to Mormon beliefs!) operating according to supernatural laws. But you haven't shown _how_ this supernatural chain of events leads to a natural event, which was the key point in question. What is the mechanism by which causes in the supernatural world can produce effects in the natural world?
It's also hard to tell exactly how you distinguish 'natural' and 'supernatural.' Why couldn't a naturalist simply subsume both 'natural laws' and 'supernatural laws' into some comprehensive description of Nature in two distinct, non-overlapping regimes, much like we've currently done with general relativity and quantum mechanics? It seems the naturalist would be fully within his rights to declare that we do not have 'supernatural causation' at all, but merely two different spheres of the natural realm.
On a related note, you had originally suggested that a lack of known purpose disqualified the design hypothesis as an explanation. And you provided just such a purpose in your hypothetical example of a legitimate supernatural explanation. But if we were to hypothesize that the divine Designer is, for instance, the God of the Bible, we would indeed have a purpose for his creation of human beings: to display His glory. We can reject this suggested purpose for the Designer, but how can we say that it is less meaningful than the purpose you offered ('Joe is really fond of the number 1 trillion') as a valid answer to the 'why' question?

Neil Shenvi Jeffery Jay Lowder • 5 days ago
"I thought that I did provide the actual mechanism for supernatural causation in my hypothetical example:"
In your example, Joe creates minds out of schmatoms. If you are approaching this problem from the perspective of substance dualism, then Joe is merely creating one non-natural entity (a mind) from another non-natural entity (a schmatom). So you haven't provided a mechanism for supernatural causation of a natural event.
On the other hand, if you're approaching this issue from the perspective of some kind of supervenience (as your original response suggested), then mental events are merely properties that supervene on physical entities. So you still haven't specified _how_ schmatoms cause or influence these physical entities. So either way, it doesn't seem like you've explained what a mechanism for supernatural causation would look like.
"The design explanations I've seen so far seem to be missing a statement about the designer's goals."
I'm sure that's not the case. Creationists always identify the Designer as the God of the Bible. But since this is a de facto objection, I don't think it's a major issue. We can always append a motive to the Designer's action which is surely no less plausible than 'Joe likes the number 1 trillion' and avoid this objection.

Neil Shenvi Jeffery Jay Lowder • 5 days ago
I'm now not sure why my previous objection doesn't apply. Since your proposal involves 'laws of supernature', I suggested that a naturalist could subsume both the laws of nature and the laws of supernature into one comprehensive description of Nature. You said that this move was impossible. But you just wrote: "The laws of supernature are such that, when schmatoms are configured into an atom interactor, certain vibrations of schmatoms have corresponding effects."
In your previous article, you defined Nature as "the spatio-temporal universe of ... [entities] which [are] the kind of entity studied by physicists or chemists." If schmatoms have the (spatio-termporal) vibrational properties you suggest and interact with the natural world, is there any reason why a physicist or chemist can't study them? After all, you have already fully characterized their vibrational properties, which is more than most string theorists can claim regarding their own field of study! But if schmatoms are, after all, natural entities, then they cannot provide examples of Supernatural causation.
So, as I and Rauss have both suggested, your definition of 'mechanism' is smuggling in the assumption of naturalism. But it's hardly fair to reject supernatural explanations as non-explanations because they fail to provide a naturalistic account of the supernatural causation!

Neil Shenvi Jeffery Jay Lowder • 4 days ago


"Schmatoms, which behave according to laws of supernature and which can causally interact with the universe despite not being a part or product of it, are supernatural entities."

Can you explain how 'schmatoms' are different than strings vibrating in 10-dimensional space or quantum mechanical wavefunctions? Why are these entities 'natural' while 'schmatoms' are 'supernatural'?

Neil Shenvi Jeffery Jay Lowder • 5 days ago

"As soon as you add a motive to the content to your hypothesis, you decrease its prior probability by decreasing its modesty and its coherence."

This seems like an odd objection, since your original claim was that design explanations ideally ought to include motives. It seems a bit unreasonable to suggest that design explanations should include motives and then penalize them for providing motives. However, I think that whatever the biblical motive loses in modesty, it more than makes up for in coherence. For instance, if God created the universe to display his glory, it explains not only the creation itself, but its habitability for life, the existence of sentient creatures, the existence of free moral agents, the existence of evil, etc...

"ID theorists go to great lengths to emphasize that the intelligent designer need not be the God of the Bible." 

Yes, it's lose-lose for ID theorists. If they supply a motive from the Bible, they are creationists. If they don't, you call them on the carpet! :-)

"You can't pick and choose and say that God's reasons are unknowable when you are responding to an argument from evil, but say that God's reasons are knowable when responding to de facto objections to theistic explanation."

This is where I think presuppositionalists rightly recognize the limits of natural theology. For a Christian who believes in special revelation, it is possible to 'pick and choose' in the sense that what God has revealed is knowable (his motives for creating the universe) but what he has not (his motives for allowing any particular instance of evil) is not. See Deut. 29:29 (which was the inspiration for Harvard's original shield) for a biblical exposition of this particular point.
Rauss Jeffery Jay Lowder • 5 days ago

Given what you've said, the most reasonable arguments you can make against immaterial explanations are that they are insufficiently modular and not as predictable as physical phenomena.

But this hardly rules them out as explanations.

At best you can say they are not excellent explanations, but that is provided you have some sensible standard of how one measures explanations.

As it is, you're begging the question against non-physical explanations. That's certainly not fair.

Rauss Jeffery Jay Lowder • 5 days ago
As it is, by demanding immaterial explanations resemble material explanations in behavior rules out by your very criteria the immaterial.
This unexamined, arbitrary criteria serves no other known purpose. We can both acknowledge this in the spirit of honesty.
After all, even the example you gave to Neil Shenvi sneaks in a reductionist account of immateriality.
Why must immaterial things be composed of something, ie schmatoms. Doesn't this presume a reductionist paradigm without evidence?

Richard_Wein • 5 days ago
Hi Jeffery,
"Compatibility is like pregnancy: a person is either pregnant or not. There is no in-between. Likewise, evidence is either compatible with a hypothesis or it’s not."
On the contrary. Surely you've heard of such ideas as the "underdetermination of theory by evidence" and the "Duhem-Quine thesis". Inferences from evidence are not purely deductive. They always involve some element of non-deductive judgement. Hence the evidence never absolutely (deductively) falsifies a hypothesis. After all, any observation could in principle be just a hallucination! So evidence is never absolutely incompatible with a hypothesis. There can only be degrees of compatibility.
Now, you might say that it's better to use some other word here, and reserve "incompatibility" for the condition where one can strictly deduce the falsehood of one proposition from another (or set of others). But I don't accept that such a restriction is required by the ordinary meaning of the word. Perhaps philosophers consistently use "incompatible" in a stricter technical sense that is restricted in this way. If that's so (and I have my doubts) then evidence can never be incompatible with a hypothesis, in this strict deductive sense.
Personally I prefer to talk about the consistency of the evidence with a hypothesis (recognising that such consistency is a matter of degree). But I wouldn't object to anyone using the word "compatibility" instead, as long as they understand that compatibility (in this context) comes in degrees.
Incidentally, many apparently precise concepts turn out to be fuzzy to some degree, and pregnancy is one of those. The concept is not sufficiently well-defined for there to be an absolutely precise moment in time when a woman goes from being not pregnant to being pregnant, just as there is no precise moment in time when life evolved from non-life, or a precise moment in time when one species evolves from another species. So there is a brief interval when a woman is neither pregnant nor not pregnant; her pregnancy state is undefined.
Mistaking fuzzy distinctions for absolute dichotomies is in my view a major cause of error in philosophy.

No comments:

Post a Comment