AstroSkeptic

Doubting Didymus and Beausoleil,

For the record, I agree creationism fails under the argument from parsimony. So in the sense that creationism is refuted on one count, one could say it is pointless to debate its merits on another. In this sense, I’ll agree the argument is futile, however would you be willing to indulge me in this? Very often old earth evidence is offered up as a refutation in and of itself. If this is correct, then we never even have to get as far as parsimony. I’m trying to at least hold a defensive line back to parsimony (where I’ll run up the white flag) by arguing that old earth evidence by itself does not suffice for falsification. Basically I’m trying to establish that creationists are reasonable in arguing that the universe had to be built “aged.”

So I do have a point and a purpose, although whether I am successful is another question. I don’t think many people allow for the necessity of a universe created “aged.” I’m trying to argue the validity of this, something I would think creationists would be interested in.

So at this point I’d like to ask a question:

Are you (a) denying the validity of the conclusion, i.e. that an act of creation of logical necessity must result in an implied past* or (b) not denying the conclusion, but arguing that it has no utility for creationists?

*Beausoleil, you said there are states with no implied past. I agree, but insofar as creationism is the claim that the universe was created in a state fit for human habitation, I submit that the created state must have had a non-zero implied past.

If (b) is the case, I would offer the following as an example of the utility which a creationist could derive:

Smith: You believe the universe was created 6000 years ago?
Jones: Yes.
Smith: Well, you’re wrong.
Jones: How is that?
Smith: We have reliable evidence that the universe is much older.
Jones: How can you establish that this particular evidence refutes my belief?
Smith: Your cosmological model predicts a universe with an age of 6000 years.
Jones: No it doesn’t. My model makes no predictions of the actual inductively implied pre-creation age of the universe, other than it is greater than zero.

I think a criticism that has been made here is to say that Jones has no predictions to offer Smith to salvage his claim. But Smith is actually the one making the claim, so is the burden of proof not on Smith? Isn’t Jones in a better position now?

Doubting Didymus

I don't know about Beausoleil, But I am granting you the conclusion temporarily and arguing for non-utility.

In your hypothetical, jones has not has his position strengthened because his position is that the creation was specifically 6000 years ago, specifically performed by the christian god, and a host of other specifics that this argument does nothing to support. What your argument does for christian creationism also works for all other forms of contradictory creationisms and thus, it can not really be used by any one of them to argue for their position specifically.

On top of this, your argument forces attributes apon the creator that would be unnacceptable to anyone trying to use it. To expound: either the creator could have made the universe to look as old as it truly is (6000 years old, in jones's case), but chose not to, or the creator had no choice but to create the universe in the way we find it. On count one, said creator is omnipotent but deliberately misleading, on count two, creator is possibly honest but not omnipotent. Thus: your argument cannot be used to support any of the creationist positions that we see in the world today.

To turn to your conclusion: that a recent creation would create an implied past comprised of a series of the most likely moments immediately previous to the proceeding moment. Upon a moments reflection, It seems to me that this does not allow the past to contain any statistical anomalies, as each moment should be the most likely predeccessor (and thus each statstical outcome in the past should have been the most probable outcome).

Does this not mean that if we examine the past and find that a given moment could have had a likely proceeding moment, but had an unlikely proceeding moment instead, the conclusion that the past was created from a string of most probable precendents is refuted? If not, you are arguing for a very strict form of determinism, that each moment could not have been otherwise.

[ August 15, 2002: Message edited by: Doubting Didymus ]</p>

AstroSkeptic

Let me take another crack at utility - You are exactly right that this argument does not help the creationist position in the sense of providing any new support. But I'm not trying to strengthen the creationist position by adding support, but rather by removing one category of potential refutation. If our degree of confidence in a claim is a function of the quantity/quality of evidence for and against, then our confidence is strengthened if the "pro" side remains unchanged while the "con" side is reduced, yes?. I'm not trying to address the "pro" side of creationism. I'll agree that if the "pro" side is null, then confidence is exactly zero. To be frank, I suspect the "pro" side is null but I'm trying to give creationism the benefit of the doubt.

So my claim for utility lies in the reduction of the "con" side, specifically a reduction in the potential for falsification. If we assume a 6000 years-ago date of creation implies that nothing in the universe should date older than this figure, then creationism is falsified because it makes age predictions that are contradicted by reality. But does creationism actually makes such predictions? If you grant my conclusion, then it does not. It makes no specific prediction with regard to age. So creationism is unfalsifiable as regards age data. But is this a problem? Creationism makes no predictions regarding the quantity of acorns we expect to observe in the universe either. So it is unfalsifiable as regards acorns and many other physical quantities we could measure. I'm not claiming that creationism is unfalsifiable in general, but just in regard to a specific category of evidence. It may make many other predictions that could be falsified (like the presence of a "flood" layer in the fundamentalist Christian version). Of course, one only needs a single falsified prediction to negate the overall claim (in which case my reduction of the "con" side is futile if the residue still includes such an instance).

So would you agree that, given my conclusion, I have reduced the "con" side? I'll grant that the "con" residue may be damning to the overall claim and hence the progress may be for naught, but at least in principle, future arguments and/or discoveries could address the residue to the point where creationism could be vindicated? (e.g. maybe geologists will one day discover the flood layer) Granted, I'm reaching here but at least in principle, removing some of the "con" side, even if what remains is still voluminous, does amount to strengthening the claim?

Actually, it could be argued that the creator would not have created a universe that would look 6000 years old to us now. To do so would require that the universe have an implied age of zero at creation. An implied age of zero would require that all causal inference would stop at the putative moment of creation. One state where this could occur is the singularity of the big bang but since any variant of creationism assumes a habitable universe at the moment of creation, such possibilities are ruled out.

The question of statistical anomalies in the implied past is an interesting one. It is certainly true that if we always assume the most probable past history, we'll be wrong some of the time. For example, the moon is believed to have been formed by way of a collision between earth and a mars-sized body as opposed to the unlikely scenario of orbital capture by the earth of the moon after it had been wandering through space. So there are many implied pasts with varying degrees of likelihood, but this conclusion only requires the existence of a single implied past - it only takes one for the conclusion to be valid. If there is more than one, we may not know which is correct (only most likely) but the salient point of the argument is that none of the alternatives is of zero duration (because a zero duration implied past means that the universe could look 6000 years old and hence I can't remove the "cons" I'm trying to remove).

Doubting Didymus

Yes, I agree with you here.

As an aside on this discussion, I am very interested in your motives for this argument. As you clearly do not personally believe in christian creationism, as evidenced by such quotes as the one above, why are you trying to strengthen its cause? Are you simply trying to hone your debating skills? (This is not a critisism of your person, but a genuine request for information)

Hypothetically this may be true. If we assume the conclusion, then creationism SHOULDN'T make such predctions. But the fact in the real world is that it DOES make those predictions, and claims that the real evidence from the scientific endevour supports a 6000 year past.

If what you are saying about creationism was true, then yes, it would not be falsifyable with those categories of evidence that it does not make claims about. This however, opens a whole other can of worms that I will soon go into.

Unfortunately for creationism, this is the case.

Okay, here is where the new worms come into it. The case for creationism is only strengthened if it abandons science, and completely stops caring about evidence. Still assuming that your conclusion is true, the argument only establishes the implied past as neccesary with respect to the creation event itself. The creation event is still left only as a possibility with precicely ZERO supporting evidence. Science must neccesarily ignore the possibility of the creation event until there is evidence for its occurance. So, even if a creation event created an implied past, we still have no evidence to suggest that the creation occured.

If you can find a brand of creationism that does not care about evidence, does not care about science, then, and only then, is that brand of creationism strengthened by your argument. (oh yes, and as I mentioned above, they must also believe that the creator is either not omnipotent or is malicious.)

I have just now re-read your argument, and I now believe it to be fallacious after all. Up till now I have been granting your conclusion, but I think there is a validity problem in reaching conclusion 2.

You reach conclusion two from premise 2 and conclusion one. To summarise this:

Conclusion one: any permissable state has one or more permissable preceeding states.

premise 2: creationism states that the universe is created in a permissable state.

There are two problems here: one is that premise 2 is not compatible with conclusion one. Any permissable state has a permissable preceeding state, yet the state at creation would not have a true preceeding state by definition, and thus one or the other must be false.

The other problem is that you make an unjustified assumtion in suggesting that conclusion 2 follows directly from C1 and P2. Although it may be true that all states have permissable precceding states, it is not neccesarily true that the created state Sc would then HAVE to imply that those states existed. For one, If state Sc is THE FIRST OF ALL STATES, then it does NOT have preceeding states at all, and so is not forced to imply them. Otherwise, your argument may say that natural laws applied to state Sc instigated the ACTUAL existence (not merely the implication) of previous states. I do not see that this is supported by our knowledge of natural laws. You propose no mechanism, nor do you suggest that it might even be possible, to have states come into existence because of a future state proceeding said state.

To summarise this point: A creationist definition of state Sc would be a state with no preceeding states, and thus premise Sc is not a state as described in your conclusion one. If no previous state exists, State Sc is not required to imply it. Alternatively, state Sc has to generate actual preceeding states through natural laws, the possibility of which is highly dubious.

AstroSkeptic

Doubting Didymus,

First off, thank you for staying with me to this point. As to my motive, it began in a discussion with a friend of mine who teaches evolution. Having had conversations with creationists before, I argued the creationist point with him, more to make sure he had a good exposure to their arguments than to hone my debate skills. I developed the present argument at this time but the roots began in an earlier debate with a real creationist who took me to task on this age issue. I had heard the universe built "aged" bit before but never had seen anything other than a handwaving justification. The more I thought about it, the more it seemed to me that it would be difficult or impossible for the universe to have been created in a state that implied no past. For example, if the creator had decided to create sedimentary rocks, then the implied age of the universe would be at least as long as it takes such rocks to form naturally. Basically, I wanted to see how far I could take this.

So I got to the point where I convinced myself that a habitable planet cannot be created without some parts of it implying a past that never occurred. Certainly that realization is miles away from demonstrating concordance with the real past as discovered by science but it seemed to give the "built aged" argument a faint heartbeat.

So from there I simply took the observation that since any state of the universe can be thought of as an effect of a prior cause, we can define a past history or set of histories to *any* putative initial state. The only exception to this being special primordial states like the big bang which have no known causes. But since such states would appear to represent uninhabitable universes, they don't apply to the creation state. But in retrospect, I'm not sure this is true and this goes back to my premise of the creation state being permissible. I think that premise is convenient for the conclusion, but probably flawed. For example, the creator could have chosen to construct mars with a giant face on the surface. Why not? If the creator didn't also construct the requisite physical laws that would imply a causal history of such an artifact, then we would have an anomaly. I'm not sure if this is the type of anomaly you alluded to, but such a state of creation would clearly have *zero* reasonably probable past histories. All science could say is "there is no explanation for this" and maybe we'd all convert over to the ID camp.

So although I can't dispute the creationist claim that some parts of the universe should imply age, I have to agree with you that this attempt to show that the universe as a whole should yield a tidy implied past just doesn't work. Premise 2 is shaky at best and as you point out, the conclusion becomes suspect under the ambiguities in the assumptions. Oh well, creationism is a tough one to argue…. I think I have a much better argument against theism in the form of an attempt to show that the god of the bible is immoral, one which I think successfully avoids the standard pitfalls in such arguments.

Doubting Didymus

I find that the best argument against theism is about lack of proof. If there is no evidence for god's existence, there is no reason to believe in the truth of god. If we accept a proposition that has no proof, then we have nothing to distinguish between such propositions, and must believe every unevidenced proposition including leprechauns.

Put formally: The only sensible position towards any positive proposition P is the unrestricted negative position by default. (otherwise the observer must accept or be agnostic to each and every possible proposition)

Only evidence of some kind, sufficient to prove to the observer the probability of proposition P, should cause the observer to reject the unrestricted negative of proposition P. (for the same reason as above)

Therefore, any positive proposition P should be rejected unless there is some evidence for it.

or:
If there is no evidence for X, Then not X.
there is no evidence for X
therefore not X

The real theism debate, therefore, should be restricted entirely to the truth of premise 2, I.E. whether there is any evidence for god.

Many theists believe there is evidence for god, and I find such theists to be logical and sensible. The debate here is with the quality of the evidence.

Some theists, however, maintain that there is no burden of proof placed on them, and that theism requires no proof. I find this to be an invalid argument. Accepting a positive proposition without evidence is illogical.

AstroSkeptic

I agree and I think where one places the threshhold of belief within the spectrum of evidential support is a function of psychological factors and the degree of personal importance of the proposition in question. Psychological factors include whether one is disposed to be more concerned with (a) accepting a falsehood or (b) rejecting a truth (type I vs type II errors). Everyone has to lean in one of these directions. Personally, I would rather filter our falsehoods at the cost of some truths than vice versa so my threshhold is set fairly high. Probably in everyday affairs theists have a similarly high threshhold which is why they can function in scientific settings. Where I think they go awry is to make the threshhold setting also a function of the perceived importance of the proposition. Enter Pascal's Wager. Somehow our decision threshhold is now a function of the stakes of the game. These people function one way in their professional lives, but they step one foot into church and their epistemological meter gets recalibrated. As far as I'm concerned, this double standard is irrational.

Doubting Didymus

The same irrational double standard applies to modern creation scientists. Many of whom are able to apply proper proceedure and review to physics and chemistry, but not to their creation theories.

Would you agree that agnosticism toward god is the exact same double standard, as there is often no agnosticism toward other, equally evidenced propositions?