You can’t know everything, therefore…

Hey world, did you know that @theealex said he could be wrong about everything he thinks he knows? #illogical

Amateur religious apologist Eric Hovind is fond of this little point, which he seems to have been hammering away at repeatedly over Facebook and Twitter recently. Despite plenty of people calling him on his bullshit, putting forward interesting points and counter-points, he seems to only ever pick on these fairly simple arguments, then twist them in such a way as to develop his own personal straw man that he can trot out repeatedly on command. It’s a useful tactic for the Twitterarti to use, because it takes far more than 140 characters to really get into this subject and call it on its bullshit.

First, what is it that Hovind is trying to say here?

In short, he claims that “the atheist” (because these religious fundamentalists work best when they take a demographic and compound them into a singular abstract entity like this) says that they can’t be sure of themselves, that everything they know could be wrong. To Hovind, a man of religion, and specifically Young Earth Creationism (YEC) religion, this simply isn’t good enough. He deals in absolutes. You’re either Wrong or you’re Right. There is no room for hedging your bets or making statements in probability. What’s more, even 99.9% certain isn’t good enough for him, 99.99% sure is poor odds, and 99.999% right is still 100% wrong. He tacks on “therefore God” on the end of this as a matter of course.

The trouble with this is simple; knowledge doesn’t work in absolutes. Every statement we make has an uncertainty attached to it. Of course, humans are emphatically terrible at guessing the level of their uncertainty in any robust form, but that’s not the point here. We just need to know that our statements present a degree of uncertainty. What I want to do here, is demonstrate that.

That the sun will rise in the east tomorrow morning is a statement that seems very sure of itself. So far, we have never experienced a day where it has failed in its duty to do so. We have little, if any, evidence to say it will stop doing so. And even the plausible suggestions as to why it would stop are far fetched. But can you discount entirely, with 100% precision and probability, that an alien Death Star will not jump out of hyperspace in the next 24 hours, destroy the planet, vaporise the sun, and ensure that the sun won’t rise in the east? If you can’t ensure that isn’t the case, and it would make for some quality entertainment to watch people try, then the statement about the sun rising in the east possesses an uncertainty. A small uncertainty, but uncertainty nonetheless. But let’s assume someone has ruled out the alien Death Star through persistence and strife as much as evidence and logic and has managed to generate a statement with an unprecedented infinite certainty. Can they also rule out a spontaneous supernova? Or can they rule out a planet destroying Earthquake? Or trans-dimensional beings absorbing the sun into a black hole? What about trans-dimensional beings absorbing the sun into a black hole while wearing red socks? No one can predict the future with sufficient precision to rule this out, so uncertainty exists in any statement we make about it.

But then we come to another example. If someone was to guess the height of the tallest man-made structure in 1921, they would almost certainly be wrong. They’d also have an uncertainty attached to their measurement. Once we’ve looked it up and factually checked it things change, our uncertainty lowers and becomes more refined, but before that when we just guess, we have an very large uncertainty. Guessing the value to the nearest millimetre will almost certainly be incorrect. Guessing the value to within 20 metres might produce some reliable answers. It’s likely that no one, off the top of their head, could get it within +/- 50% in all probability. So such a statement, such a guess, has an uncertainty about it also.

All statements are uncertain in this way. But this is where the concept of the continuum fallacy comes in, and where Eric Hovind’s point about the atheist “not knowing anything” falls foul of a significant equivocation error. All statements have uncertainty, but that degree of uncertainty is not the same across all statements, and treating them as such is simply wrong. Indeed, that itself would be a statement that is wronger than wrong.

Unfortunately, any atheist or science advocate trotting out this “but we could be wrong” statement isn’t doing themselves any favours for this exact reason. And hence why Eric Hovind is keen to jump on it; it’s overly-simplistic and plays into the hands of people who want  some of our most well-defined knowledge to be wrong. The “we could be wrong about everything” statement is presenting a weak argument that lacks the subtly to how knowledge actually accumulates and how the theories we have on how the world actually develop. Theory isn’t split into a false dichotomy of Right or Wrong, each idea is individually judged on its ability to make predictive statements. Newton’s laws of motion, not having taken into account relativistic corrections nor any quantum mechanical concepts, are wrong. But they still work to make predictions. A car travelling along a road obeys Newtonian mechanics quite well, a planet less so, and electron even less so. Does this make classical mechanics wrong? No, it simply makes it less accurate.

Isn’t this the same as wrong?

It depends on how you want to define “Right” and “Wrong” given that the real world presents you with no such distinction.

You can define Newton’s Laws as “wrong” if you like, after all, they do lack certain features we know to be required for a precise description of the world at other scales. Nothing is technically stopping you from doing so and defining these words as you like; so long as you use this meaning consistently, and don’t try to make any inference beyond that meaning. You can call it “wrong”. You can say quantum mechanics is “wrong” by the same principle. You can say evolutionary biology is “wrong”. But it would be trivial unless you could find something that isn’t “wrong” by this definition. If a statement that is inaccurate or imprecise because it lacks infinite certainty, then every statement is wrong. In would be better to split the world of ideas and predictions into a more realistic format, where they’re judged according to their individual accuracy and precision, rather than lumped crudely into two arbitrarily defined buckets of “Right” and “Wrong”, but it’s unlikely the human brain can take that sort of categorisation. We like all-or-nothing binaries too much for that to work on a day-to-day basis. 

By insisting on splitting the world into a dichotomy of Right and Wrong theories (factual theories, lest anyone get confused with moral relativism), you simply produce a situation where you have only one category, Wrong. A single category to cover every proposition, idea, hypothesis and theory, is useless and lacks any meaning because it can’t whittle down everything into something. It would be as if declaring everything in the world to be “small” simply because we can raise a distance measurement to any arbitrarily large power. Such a distinction wouldn’t serve us at all and would place the size of a field mouse into the same category as the size of the planet, and the same category as the size of an atom. Similarly, a dichotomy of Right and Wrong doesn’t serve us at all because everything now lies in the category of “Wrong” because such categorisation refuses to accept the realities of how uncertainty works.

This is the world as Eric Hovind has to present it. Everything is either Right or Wrong, with no room for uncertainty. What’s worse, is that any uncertainty is tantamount to being Wrong. Not only is this not reflected in reality, but even Hovind’s own little world produces completely meaningless distinctions. But it serves his purpose, he wants to evangelise his own Good Word and pretend that the realities of uncertainty don’t exist, because that would be difficult.