Sep 13 2011

Juicy Post-Hoc Reasoning

Last week I wrote a post about Billy Meier and his many decade claim for ongoing alien contact. To me it stands as one of the best examples of the human capacity for rationalization. Meier’s clumsy attempts to hoax cheesy flying saucers, photos of aliens, and other bits of evidence are pure gold for skeptical analysis. Demonstrating that his “evidence” is easily explained as hoaxes is fun for anyone interested in doing such investigation. For me the more interesting bit is the psychology of those who actually believe Meier’s claims.

Extreme cases like the Meier case provide very useful teaching moments. They are similar to a physician calling students into an examination room to witness an advanced stage of an illness – in the advanced state some signs and symptoms are likely to be more easily appreciated. They can then serve as a dramatic demonstration of the core features of the disease.

In response to my post last week, commenter Jamesm responded:

To respond to your first point stating that the beamship was probably attached to the tree, the “some reason” for this was as follows:If the Plejaren allowed Meier such close up photography of this craft without any possible means to reject it (such as a nearby trees or other objects) then it would make it UNDENIABLE in the mind of the common man and this would result in them directly and significantly interfering in the evolution of a developing pre-interstellar civilisation (that’s us). In fictitious Star Trek terms, that’s called violating the prime directive and is not permitted. The case is apparently the same with the real Plejaren Federation.

I had predicted that someone motivated to believe Meier could find some reason to explain why the UFO appears to be attached to a tree. James has now nicely accommodated my prediction with a perfect example.

The problem with James’ claim is not that it is incompatible with the evidence, but rather that it is an example of post-hoc reasoning (also referred to as special pleading). Humans are very good as seeing patterns, making connections, and coming up with explanations for phenomena. Almost anything can be explained in retrospect, and it only takes a couple of apparent connections to make such explanations seem compelling.

That is actually the fatal flaw of such post-hoc reasoning – it can be used to explain anything. Therefore, the fact that someone can invent a post-hoc justification is not predictive that the explanation is actually true.

In other words – we tend to be naively impressed with the fact that an explanation is available. We tend to assume that a phenomenon would not have an explanation, or alternative explanation, if that explanation were not true. We therefore find it satisfying, and therefore compelling, when such an explanation is available.

However, the human capacity for invention and pattern seeking is profound. We can find an explanation for anything – and so the availability of such an explanation should not be compelling at all. The ability of Star Trek fans to invent reasons to plug the gaping plot holes or technological anomalies that often crop up on the show is an example of human creativity in this direction – although this is done for fun.

James’ comment is a perfect example of when such reasoning is used seriously. What he is saying, in essence, is that the Billy Meier UFO phenomenon looks like a hoax because the aliens deliberately allow him only to have the sort of ambiguous evidence that is compatible with a hoax.

This rationalization is not unique to Meier. The UFO community has been using this one for a while – the aliens are revealing themselves in bits and pieces to prepare us psychologically for the time when they openly reveal themselves. They prevent anyone from obtaining smoking-gun proof – only glimpses and ambiguous evidence are allowed, so that plausible deniability can be maintained.

This way the evidence is there for those who are ready to see, but there is no unambiguous proof that would convince a skeptic.

This type of reasoning is insidious because the explanations that are invented post-hoc cannot be disproved with evidence. That is because they were invented after the evidence specifically to explain it, not based on any prior logic or plausibility. So of course it fits the evidence – that is the essence of post-hoc reasoning.

Post-hoc explanations fail for reasons other than the evidence. The first is that they are not validated by predictions. The ease with which we can retroactively explain evidence is exactly why science relies heavily on making predictions – which is a much more reliable test of the validity of a theory. For example, astrologers are very good at explaining why something happened based upon the astrological signs. They are completely unable, however, to use the same astrological signs to predict what is going to happen (better than informed guessing).

The second failure of post-hoc reasoning is Occam’s razor – the rule of thumb that when multiple theories are compatible with the evidence, the one that introduces the fewest new assumptions should be preferred. The reason for this is not that elegant solutions tend to be correct more often than complex one  – they may be,  but the world can also be a complex place. Rather – Occam’s razor is useful because complex explanations are often rigged with special pleading to invent specific explanations for specific pieces of the evidence.

It is that post-hoc rigging that we need to be very suspicious of.

Let’s bring this back to the Meier case. We can predict, based upon sound reasoning, that if Meier’s claims were true then over time compelling evidence would emerge to support his claims. If, however, Meier’s claims are false and he is perpetrating a hoax (or perhaps an elaborate hoax is being perpetrated on him) then we would expect all the evidence that Meier produces to be compatible with a hoax.

Over the years every bit of evidence that Meier has produced is compatible with a hoax (and a fairly childish one at that). He has been caught essentially red handed numerous times. Each time he invents a far-fetched special explanation for that one case. I recounted some of this in my previous post, and here they are with some more:

– The UFO base looks like a garbage can lid found in his house because of some crossed ESP signals
– Another UFO looks like a model dangling from a string because that’s just how their “beam ships” move
– The UFO is hugging a tree because the aliens want it to look like a hoax
– Men in black replaced the photos of the aliens with human look-a-likes, and Meier forgot that he was given this information years earlier and had to be reminded

Or (here is where Occam’s razor comes in) it’s all a hoax. Then we don’t have to introduce aliens, ESP, the Men in Black, time travel (Meier claims he was taken back in time but his picture of a Pteranodon was demonstrably taken from a book), and all the rest.

Such reasoning should always be a giant red flag – whenever someone argues that the circumstances have conspired to make it look as if a claim is false, when in fact the claim is true, they are likely special pleading. For example, the argument that God made life on earth to appear as if it evolved is special pleading. The argument that Bigfoot is psychic or can become invisible and that’s how he evades capture or unambiguous photos is special pleading. The notion that skeptics give off anti-ESP mojo is special pleading.

These are blatant examples, but they illustrate the basic principle. The trick is to be aware of the more subtle and reasonable-sounding examples of the same faulty reasoning that we use in our every day lives. We need to put a “special pleading filter” in place, to monitor our reasoning for convenient post-hoc explanations.

Our critical thinking skills would also benefit from metacognition that tells us not to be impressed with pattern recognition. When all the pieces seem to fit into place, we have to ask ourselves – could this all be a coincidence? Were the pieces chosen from a much larger set of facts, and were their significance chosen only after the correlation was detected? This can be very counterintuitive – we want to be impressed with the patterns we detect. But they are mostly illusions, albeit compelling ones.

46 responses so far