Extraordinary claims require extraordinary evidence

The title quote is a popular one with skeptics, and is often attributed to Carl Sagan in his Cosmos series, though Wikipedia reckons that Marcello Truzzi and Pierre-Simon Laplace both said something pretty similar first. I think I prefer another assertion along the same lines, known as Hume‘s maxim, which states:

That no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous than the fact which it endeavors to establish.

This, I think, expresses what Sagan and the others were getting at, perhaps a little less succinctly, but also more clearly and unambiguously. It’s closely related to Occam’s Razor, or the law of parsimony, but it deserves some elaboration as to its exact meaning. It’s not really fair if “extraordinary evidence” is allowed to become a vague and unattainable moving goal-post for any phenomenon not already established.

If you met me at a party, and I introduced myself as Greg, you’d have little reason to doubt me on this point (assuming you don’t already know that that’s not my name) and probably assume it to be true without a moment’s hesitation. Why would I be lying about my own name, to someone I’m meeting in some casual social situation? It’s hardly an extraordinary claim, so extraordinary evidence is not really required – my word is probably enough to convince you.

If I were to further claim to have made a considerable fortune by hiring myself out as a private contractor to service the sexual needs of Natalie Portman, I’d probably lose a smidgen of credibility. The claim is, after all, on the incredible side, and the evidence I’ve provided is rather lacking, so you’d be reasonable to assume that I’m probably attempting to live out some personal lonely fantasy.

Looking at it in Hume’s terms, it’s unlikely to be more than a very minor miracle for me to give you my real name. I could be lying for some reason, but who does that? It’s not impossible, but marginally more miraculous, so you’re probably safe assuming you do know my name.

However, my earning any money at all as a man-whore to today’s Hollywood starlets would make raising Lazarus from the dead look about as impressive as stealing a kid’s nose. And all you’ve got is my testimony, which ain’t much. It might be odd that I’d tell such a brazen lie – most people aren’t mentally ill, and don’t just make up obvious nonsense to try and impress random strangers – but any explanation is more plausible than that I’m telling the truth. The falsehood of my testimony is hardly miraculous at all in comparison, so there’s probably not much point asking for advice on getting into my line of work.

We’ve all seen people lie for personal gain, or be wrong but sincere in their efforts to convince us of something they believe. I can’t even be bothered to find any links or references to corroborate it right now – that’s how obvious it is that people might easily claim to believe all sorts of untrue or incorrect things, no matter where they fall on the scale of honesty. If someone says, for instance, that Elvis beamed down into their bedroom last night from his new home planet, then there’s nothing very miraculous in suggesting that they’re either mistaken or lying.

If, however, the King then attends a major press conference announcing his new comeback tour, dazzles millions with a series of concerts around the world, and provides several independent scientific agencies with samples of the technology he uses for his interstellar travel, then we might start to be convinced. There would come a point where the “everyone’s just making it up and/or experiencing a mass hallucination” idea falls apart.

And the principle is often applied in actual, non-Elvis-based science, too. When Einstein first published his ideas about the curvature of space-time, gravitational time dilation, and other aspects of general relativity, it was pretty extraordinary and/or miraculous stuff. The evidence initially in its favour couldn’t have been called extraordinary, and for it to have been false didn’t seem to require anything more miraculous than a German scientist with some wacky ideas.

But then the evidence started coming in to support it, observations which Newtonian physics alone was unable to explain. When light was observed to bend around a massive object like the sun in just the way he predicted, in a way inconsistent with classical predictions, the falsehood of Einstein’s testimony became more and more unlikely. So these ideas, which started off sounding improbable and being rejected by most scientists, are now widely accepted. The extraordinary evidence has been provided, the miracle confirmed, and science has moved forward.

It’s a shame the ufologists, psychics, homeopaths, and so many other paranormal experts haven’t had similar luck.

The controversy around this principle comes, I guess, in making a judgment call on just how miraculous something is, and how reliable the testimony in comparison. This is bound to be at least partially a subjective measurement, but we can try and minimise how arbitrary and personal some of the decisions are. We should want to avoid setting our bar for “extraordinary evidence” so low that we become gullible and easily taken advantage of, and to do that well it’s worth understanding some of the ways that people can get things wrong. This is the kind of thing I’m planning to keep listing under ‘Wrongness’, near the bottom of the Skeptictionary list on the right there: the various kinds of faulty reasoning and misleading experiences that we can easily fall prey to, thanks to the unreliable nature of our squishy brains what do all our thinking.

If someone wakes up in the morning with a vivid memory of lying paralysed in their bed a few hours ago while a demon hovered over them, they may find this to be sufficient evidence that they were visited by something horrible from the pits of Hell. If they later read about sleep paralysis and find out about our understanding of the brain activity that can bring about these sensations, and the tests that have been done to examine what causes it and what might stop it, will they feel reassured?

Some people who go through this aren’t reassured, whatever they learn. I’ve never had it happen to me, but I understand it’s a pretty scary experience, which I can imagine feeling very real and leaving a lasting effect. But let’s take a less contentious example. I doubt many children have ever been seriously impressed by that trick where it looks like you’re detatching your thumb, but if they were, they were probably successfully disillusioned after being shown how it was done. Once you’ve seen how it can easily be faked, you assume that’s what was happening. It doesn’t make much sense to keep believing that Dad is just like Lego.

The same principles are at play in both cases. For a claim as extraordinary as being visited and imprisoned in your own body by minions of Satan, is “I remember feeling it happen, at around 4am, right after I remember running naked to catch a school bus” sufficiently extraordinary evidence? Is there anything particularly miraculous about the falsehood of the testimony – the idea that it might not have really happened the way you remember, given the other freaky shit we know your brain sometimes comes up with when you’re half-asleep?

I’ll try and go into some of the examples I’ve mentioned here in more detail later. For now, just take away the fact that, if you’re going to make a radical claim that undercuts vast swathes of established scientific understanding or invokes new and unimagined forces beyond the ken of man, you’re going to have to make a convincing case that there’s definitely something new going on the way you describe it, before we stop laughing at you and decide to completely re-evaluate our understanding of the world. It’s just basic skepticism.

Read Full Post »

Is evolution “just a theory”?

(Spoiler alert: no.)

One of the criticisms most often levelled at the theory of evolution is that it’s “just a theory”. The clue’s in the name, after all. So if we’re not really all that sure about it, some people tell us, we ought to at least consider some alternative ideas. If it’s “just a theory”, then it’s apparently not yet a fact, and so to insist that no other explanations should even be considered is unfair to other worldviews, most notably Creationism (and/or Intelligent Design).

There’s a number of things wrong with this.

Let me clarify, though, that my problem isn’t with the labelling of evolution as a “theory”. It may be the case that scientists aren’t always the most effective people to do their own PR work, but their image problem isn’t so hopeless that such a common phrase as “the theory of evolution” isn’t even accurate. It is a theory.

My problems – two of them, I suppose – are about the word “just”.

First of all, saying that evolution is “just” a theory is kinda like saying the United States “only” won thirty-six gold medals at the 2008 Beijing Olympics, or that Sylvia Browne has given “hardly any” desperate parents utterly false information about the whereabouts of their missing children. However much you try and play it down, it’s still quite a lot.

A scientific idea about how stuff works doesn’t get to be a theory the moment someone stands up on a box on a street corner and starts shouting about it, or even the minute they get a paper about it published in a respected journal. A theory is a complete model, which describes a phenomenon, and which has stood up to testing against actual data. It’s a word that’s casually tossed around a lot out in the rest of the world, but in science, theories are tough sons of bitches that have gone through the mill. It’s not a term that denigrates anything; if anything, it’s quite a badge of honour. (I’ve written more about the scientific method before.)

My other problem with this statement is that it’s not even true. Evolution isn’t just a theory. Gould had a great line about this:

Evolution is a theory. It is also a fact. And facts and theories are different things, not rungs in a hierarchy of increasing certainty.

This is to do with those pesky differences between how scientists use words, and how they’re more commonly understood. (What’s with scientists thinking they can just use words to mean what they want them to mean, anyway? Next they’ll be telling us a parsec isn’t really a measurement of time, or something.)

Colloquially, if you’ve no idea of the answer to something, you might take a guess. If you’re got a bit of an idea, and are trying to impress someone who’s easily impressed by slightly long words, you could have a hypothesis. Pretty much interchangeably with that, you may instead come up with a theory. If you’re more sure you know what’s going on, you might make an assertion, or an allegation. And something that’s not even up for discussion any more, because nobody has any doubt about what’s really true, is a fact.

But in a scientific context, these terms don’t line up in a hierarchy of increasing certainty like that, any more than an apple is more definitely a fruit than a strawberry, which itself is better than a pear. A theory is a good thing to have, a model for explaining stuff; it doesn’t mean you’ve still got a way to go before you’ve “proved” anything. Facts are useful, and can be gathered out in the real world as data is observed, but facts are things that need to be explained, and predicted, by some overall vision of what’s going on, and an understanding of what they mean. That’s what a theory is for.

One of Wikipedia’s pages on evolution has a lot more on this, and some quite gripping drama on the discussion page, too.

Read Full Post »

Pareidolia

:-O

Shock! Horror! Amazement! All obviously and immediately recognisable from two dots, a straight line, and a circle. Quite clearly, it’s a drawing of a human face.

Except, honestly, it doesn’t look anything like an actual face. Nobody’s face remotely resembles that tiny and grossly simplistic handful of symbols. People’s eyes have curved edges surrounding white-ish pools inscribed with a flecked ring of colour containing a central circle of black. If any of these details is missing or misplaced or misshaped, even in a relatively minor way, it would make for an unreal or unsatisfactory image of a human eye. If somebody’s nose was really nothing more than a thin straight line directly down the middle of their face, you’d assume they’d had a terrible accident. And nobody’s mouth really just forms handy geometrical shapes to convey easily recognisable emotions, without the rest of the face changing by a single pixel.

Clearly, those smiley faces which abound across the length of the intertubes look nothing like actual faces. But they do to us. They leap out at us and into our consciousness, often long before we register the bits of punctuation and alphabet that they’re made of, giving us the impression of seeing another human person :-O looking surprised, :-D grinning happily, ;-) giving us a cheeky wink, o_O boggling at something that disconcerts or troubles them, D: wailing at something distressing, and so on.

We readily see human faces in these shapes, because our brains are particularly good at recognising them, spotting the shapes and fitting together the patterns that comprise the visage of another person like us. There’s science about it somewhere. Probably on the internet. It shouldn’t be surprising, then, that our synapses are sometimes a little too keen to slot things together into familiar shapes and patterns, and can end up excitedly pointing out what looks like a human face, when actually all we’re looking at is a bunch of crap which just happens to have some features in roughly the right places to set our face-sensors twitching.

This is pareidolia. It doesn’t only apply to faces or human-looking shapes, but that’s perhaps how it’s most commonly perceived. Sometimes, in a cluster of chaotic visual or aural stimuli, patterns of some sort will leap out at you and seem significant.

It’s a fun game to look at clouds and decide what the vague fluffy shapes look like to you, but you don’t generally interpret this as a sign that anyone’s sending you a secret message, which might be something to do with a tiger… or maybe it’s more like a racing car. You should expect recognisable patterns to crop up amidst random stuff from time to time, if you’re looking for them in a place where lots of randomness is floating about, like among the clouds.

Sometimes, things just look like faces, for no particular reason. It would be weird if things didn’t randomly bunch together into a sort of smiley shape once in a while. When they do, you have to think whether it really means something – whether some external force has really shaped reality in some impossible way to create this pattern for some unfathomable purpose – or whether it might just be a fluke, a trick of the light, a lucky smudge. Is it impossible that it happened by chance?

Popular examples of pareidolia include the face on Mars, and reverse speech. Both merit a future entry, but those links cover it pretty nicely for now.

Read Full Post »

There are no coincidences

This doesn’t directly relate to any particular skeptical topic, but it seems to come up somewhat obliquely in reference to many different ideas, so I thought it’d be good to gather a few thoughts into one place.

It’s a common mantra among some people that “There are no coincidences”. What this generally means is that, whenever something seems serendipitous, or appears to have come about by chance in a particularly orderly way, or contains an unlikely pattern reminiscent of something else, or is in any other way quirkily coincidental, then you can bet that there’s something deeper going on, something we’re not seeing but which has caused things to align themselves in such a pattern, something that deliberately made things this way. Anytime something seems oddly out of place, or circumstances are just a little too “convenient”, you should be suspicious, and try to find out what’s really going on under the surface. There are no coincidences.

My position, however, is that if there really were no coincidences, this would be the most phenomenal coincidence imaginable.

Well, think about it. Imagine there are no coincidences. Not a single one. If you ever get talking to someone at a party, and find out that you both have the same birthday, then the two of you must have been brought together for a reason. It’s impossible for anyone to ever just stumble across any one of the millions of other people who were also being born around the same time that they were (or even on the same day on a different year), simply by chance. It could only happen when some underlying force makes it happen.

Of all the thousands of fleeting thoughts that pass through a person’s head each day, nobody could ever find any possible correlation between the vague and unprompted recollection of a person or place they haven’t thought about much in a while, and a phone call or other physical reminder of that same person or place, unless the thought was deliberately put in your head by some force that knew you were about to get that phone call. It could never happen that your idle musings just happened to overlap with reality by chance.

Out of the entire human history of bits of fruit that have gone slightly mouldy, or bits of cheese that have got burnt, not a single splotch of scorched dairy or unripe tomato could ever possibly have naturally curved itself into a pattern that sorta kinda looks like a person’s face a bit. You only need to look at the ubiquity of emoticon smileys to realise how low the human brain’s threshold is for spotting another human face (more on pareidolia soon), but even so, any simple pattern with elements that seem to remind us of facial features, like a couple of dots for eyes and a bit of a curve for a mouth, could never ever have simply come about by a random process of creating splotchy patterns, like burning a bit of cheese or letting mould grow in a tomato. It must have been placed there for some deliberate purpose.

Nobody could ever stand in front of an audience of people who are all hopeful to hear some good news about their dead relatives, and just happen to guess that the name John might mean something to someone. If anyone in a crowd of several hundred people is called John, or knows someone called John, or ever did know someone called John, then there’s no way it could just be a coincidence that a psychic happened to pick that specific name. Clearly whenever someone has such knowledge, it can only be through true spiritual intuition and guidance.

However often millions of people check their bank balances, at no point will any of them ever discover that the number happens to be in some numerical sequence that stands out to us, like £666.66, or £1234.56, or any of the other uncountable ways we could find patterns in such a string of digits. It’s just not plausible that a massive sample of random numbers constantly fluctuating up and down in varying degrees could ever land on any of the possible results that look pleasing to us. It must be portentous of something.

Oh, and those Rorschach inkblot tests? No way is it a coincidence that a bunny with a chainsaw was staring right at me in three consecutive cards. Whatever I’m getting an image of amidst the random visual “noise” of the inkblot, someone must have put it there deliberately.

Coincidence can clearly take an extremely wide range of potential weirdness, and at the mundane end we’re unsurprised to bump into them all the time. I don’t freak out about cosmic synchronicity every time I meet someone who shares my first name; it’s a pretty common first name, so obviously it’s going to happen a fair bit. Less likely is that I’ll find someone who shares my birthday, but 1 in 365 (ish) still isn’t exactly suspicious, and you only have to put 23 people in a room together to make it more likely than not that some of them will share a birthday.

On the other hand, there can be some big coincidences, that really do seem too good to be true, too profound and improbable to be simply down to chance. Lightning is a perfectly natural phenomenon, but if a bolt of it came crashing down through my roof and vaporised me the very moment after I declared “I swear, I was only holding that porn for a friend, and may God strike me down where I stand if I’m lying,” then this might raise even the most skeptical of eyebrows. If such an unlikely coupling of events did coincide, then we might really be persuaded to look for a deeper cause, which could have brought things about more plausibly than plain luck.

But the very fact that you can see a difference between the nature of these two situations – the former of “Hi, I’m Steve,” “Hey, me too”; the latter an impeccably timed lightning bolt – clearly demonstrates that some kind of judgment call has to be made here. We all expect some level of freaky coincidence to just happen, and we have every reason to expect the random noise of any media to produce some unlikely-seeming patterns now and then. Look far enough into the digits of pi, and you might find your telephone number. Flip a coin often enough, and eventually you’ll get ten heads in a row. Nearly every week, someone in the country overcomes millions-to-one odds and wins the lottery. And if any of these is simply a silly example, and your personal preferred coincidence is much less frivolous, then you’re performing some sort of evaluation to determine the weirdness, and to assess just how implausible the “coincidence” explanation is.

If you find conspiracy in every purported coincidence – literally any time two or more artefacts “coincide” – then you’ll never have the time to notice anything else. What’s important is to have enough of a mathematical understanding to distinguish genuine weirdness – where random chance truly becomes less likely to have caused something than deliberate intent – from the times when slightly kooky stuff just happens in an entirely expected way. We all make those judgment calls. It’s just a matter of whether you’re sufficiently informed and equipped to make them well. Actually looking at the odds and figuring out how suspicious to be of something is always better than trusting your gut and going with what feels more likely. Just ask Monty Hall.

Read Full Post »