The argument from minimal facts for extraordinary/miraculous events

Here is a template for an argument from the minimal facts used for example, to argue for the resurrection (see Gary Habermas here for example).

1)      These facts are agreed on as our starting point.
2)      There is a variety of explanations of these facts, including the explanation that [insert preferred extraordinary and/or miraculous event E] happened
3)      All of these explanations fail to have the explanatory scope or power for all of the facts, apart from the explanation that [E] happened.
4)      There is no compelling reason to exclude the explanation that [E] happened.
5)      Therefore (probably) [E].

This is a an interesting schema, I think. You find it employed to justify a wide variety of "extraordinary" claims. I am compiling a list of examples, so if you have any, do please let me know (include as a comment, with web link, or whatever). Quotes or clips would be particularly useful.


Justin Vacula said…
An older piece I wrote in response to some minimal fact approaches from Habermas. I found that many of the 'minimal facts' really have nothing to do with the claim (i.e. people fervently believed Jesus was divine, death caused despair, a church grew...)
Stephen Law said…
Thanks Justin I'll read that...
Rasmus said…
The first thing that comes to mind to follow that scheme (besides the resurrection), is the 9/11 conspiracy, though I'm sure you've thought about that one already.
J. J. Ramsey said…
Offhand, I don't see anything wrong with the template itself. It's just that the arguments based on that template have so far failed to satisfy one or more parts of the template. I'd say that most such arguments fail at least at point #3, and often they fail at point #1 as well.

I suppose that one can argue that miracles run afoul of point #4, but I'd tend to argue that the explanatory power of miracle claims is compromised both by the nature of a purported miracle as a rare event and by the poor track record of miracle claims in practice, which puts things back at point #3.
AIGBusted said…
I think the problems with this are two-fold:

1) The "facts" that believers in the extraordinary cite are often not facts (i.e. the empty tomb of jesus is not an established fact, many scholars disagree with it and there are holes in the arguments for it).

2) The logic of the argument is poor. Could there be a natural explanation that no one has thought of yet? Of course. And we would probably not believe other extraordinary events, like "aliens built the pyramids" even were it the case that no normal explanation was known to us. So we've got a problem there as well.
Stephen Law said…
Here's a nice example...

VinnyJH57 said…
The "minimal facts" approach is the kind of thing that conspiracy nuts use. The 911 Truthers cherry pick some video clips and bits of testimony and then assert that the only valid explanation for the fall of the Twin Towers is the one that explains those few bits of evidence. Not surprisingly, when only those bits of evidence are considered a government conspiracy sounds much more reasonable than it does when all the evidence is considered. When all the evidence is considered, the case for a terrorist attack is so overwhelming that the few bits of evidence cited by the Truthers can be safely ignored as anomalies.

One of the facts that I would use as a starting point is that dead people stay dead. It is supported by much stronger evidence than any of Habermas's facts.
Anonymous said…
You should wathc the WIlliam lane Craig vs Bart Ehrman debate and the WLC vs Arif Ahmed debate and the Arif Ahmed vs Gary Habermas debate.
Tony Sidaway said…
It offends parsimony, is the main problem I see. As we gain experience in resolving questions that were previously the realm of supernatural explanations we also gain confidence in the power of investigation. A parsimonious approach observes that there are many problems that were eventually resolved by acquiring more information, so it's okay to say "we cannot explain this yet, let's gather more information." Assuming supernaturalism is not necessary and can be eliminated without losing anything of significance.
Angra Mainyu said…
Hi, Stephen,

My couple of cents:

Question: Is a very low prior probability of [E] a compelling reason in the sense intended in premise 4?

If the answer is “yes”, then that alone blocks arguments for the resurrection and the like. People do not do that – not even religious leaders who claim powers.
If the answer is “no”, then the argument is improper, because it fails to consider the priors.

For example, in the past, nuclear fusion was not known, and there was no known way in which the Sun could emit as much heat and light as it did and does.
Some possibilities were considered, but dismissed. But if someone had said that E: = “Heat and light were coming from the Sun because an unembodied intelligent being is sending it our way”, clearly that would not have been a good argument for the existence of an unembodied intelligent being, even granting that such claim is coherent.
Concluding that probably [E] would be improper even in the absence of scientific knowledge, or history of those (purported) explanations.

Moreover, it seems for any [E] in the context of those arguments, there is some alternative [E1] not difficult to construct, incompatible with [E], and such that [E1] also would (if true) explain the observations, so it’s not clear how condition 3) might be met.

For example, if it’s offered as [E] that the creator of all other concrete beings did X, an alternative [E1] is usually that a powerful creature did X – not a good alternative, but just to illustrate the point.
Granted, someone might claim they can properly rule out all other options that have been considered. But the burden would be on them – and of course, there is the previous problem I mentioned, so I reckon the argument would fail anyway.
Stephen Law said…
Much of this fits the general rubric too.
Tony Lloyd said…
There are two ways that a theory T can fail to explain evidence E.

1. T can predict not E, it can be incompatible with the evidence or

2. T can fail to predict either E or not E, T is perfectly compatible with but not explanatory of E

The “minimal facts” approach seems to me to only use “2” to weed out rival theories. Two examples (one nicely outside any contentious areas):

Big Bang (BB), Steady State (SS) and the Background Microwave Radiation (BMR)
BB and SS were theories that both successfully explained the same evidence and neither had been refuted. The only way one could prefer one over the other was on metaphysical grounds. This changed when BMR was discovered. BB predicts BMR. SS does not predict BMR, and neither does it predict the absence of BMR. SS is compatible with both the existence of BMR and the non existence of BMR and, so, fails to explain BMR. As a result of the discovery of BMR, SS was rejected as it (now) had inferior explanatory power.

Darwin (D), Intelligent Design (ID) and homology
D forms part of an explanation for why the the bat’s “wing” looks exactly like a deformed hand (and our hand looks like a deformed paw) etc. ID (as its proponents often tell us) is perfectly compatible homology, the designer simply decided to use similar designs in similar places. But ID is not explanatory of homology as the absence of homology is also compatible with ID. So homology provides a minimal facts type argument for D.
Anonymous said…
Uhmm... B theory of time and general relativity go hand in hand together. Theists have been trying so hard keep to their presentist ways but sadly for them B theory of time is the greater of the two theories of time and as such the universe is eternal. It doesn't matter if the big bang model or steady state model is true.
Zach said…

Fact: Many people claimed to see the Hindu god Ganesha literally drink their milk offerings.

For a time there was no plausible mundane or naturalistic explanation

However, scientists later concluded that capillary action was a viable explanation.

Zach said…

Fact: People observed new living creatures being produced from "non-living" (i.e. meat) substances.

For a time there was no way to explain this emergence of life other than spontaneous generation.

Was later falsified by Pasteur.
Jess said…
Thanks to you Stephen I'm watching that Gary Habermas vid. Very interesting...
S Johnson said…
It's step four that seems to be the tricky part. If you believe that science has discovered philosophical materialism is currently the only tenable ontology, violations of natural law are justifications for exclusion. Most arguments for E would then break down at that point.

But as I understand it, scientific realism and philosophical materialism are unacceptable positions. And given a skeptical epistemology it's hard to conceive any grounds for ruling much of anything out.

The thing about probabilistic arguments in practice is the ease it is to massage the percentages. And then even if you somehow refuse to accept the possibility that some one can choose to gamble on the lower probability, the opponent can argue that one can always hedge the bet.

I'm pretty sure you can usually demonstrate unsound arguments on this scheme. Except that philosophy seems to accept only logical validity as relevant.

It seems Habermas et al. win?
S Johnson said…
PS And forgot, this works for all ESP and UFO.
CaptainDisguise said…
Islamic apologists have a similar line of reasoning to defend the doctrine that the Quran is the literal words of God. (see, for example,

The general line of argument starts with the apologist quoting Muslim scholars who (doctrinally/dogmatically)hold the view that the Quran is extraordinary. They claim it is beyond the capacity of humans so any explanation other than a miraculous one is dismissed. Then, it is concluded that the Quran is divine.

This argument is weaker than the argument from the minimal facts since the "fact" of a extraordinary is Qur'an as a starting point is not universally expected and is merely Islamic dogma.
Joseph said…
I'm a bit curious. Since you mentioned Habermas, have you watched his debate with Arif Ahmed regarding the resurrection? Ahmed's approach was to basically agree - even if just for the sake of the argument - about the so called facts regarding the resurrection, but demonstrate that even then, even if we agree with all of the claims that the proponent of the resurrection makes, we have no good reason to conclude that Jesus rose bodily from the dead.

Here's the debate, if anyone is interested:
Edward Ockham said…
"How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?"
Anonymous said…
I do not have a specific example, but I think that one can not explain the unknown by attempting to appeal to the unknowable (i.e., supernatural).
If a miracle is appealed to, then given all possiblities regarding the miracle one may assume on a continuum that any miracle may, in and of itself, be impossible. Since E may be impossible, then E is less likely than the most improbable but possible.